Functional Sensibility of the Hand in Leprosy Patients

Transcription

1 Functional Sensibility of the Hand in Leprosy Patients Wim H. van Brakel a, C. Marleen Kets b, Monique E. van Leerdam b, Ishwar B. Khawas a and Khadga Singh Gurung a a Green Pastures Hospital, P.O. Box 28, Pokhara, Nepal b Haarlemmermeerstraat 155 1, 1058 JZ Amsterdam, The Netherlands Key words: Assessment of Nerve Function in Leprosy, Peripheral Neuropathy in Leprosy, Functional Sensibility, Semmes-Weinstein monofilaments, Moving two-point discrimination Correspondence address: Dr. Wim H. van Brakel c/o INF Leprosy Project P.O. Box 5 Pokhara Nepal Brakel@npl.healthnet.org

2 SUMMARY Aims: To compare the results of Semmes-Weinstein monofilament testing (SWM) and moving 2-point discrimination (M2PD) with four tests of functional sensibility: object recognition, size discrimination, dot detection and texture discrimination. Study design: Cross-sectional comparative study. Study population: Ninety-eight leprosy in- and outpatients at Green Pastures Hospital in Pokhara, Nepal. Methods: Patients were tested with each of the above tests and the results were compared to see how well they agreed. Using the tests of functional sensibility as reference tests, we examined the validity of the SWM and M2PD as predictors of functional sensibility. Results: There was definite, but only moderate correlation between monofilament thresholds, M2PD thresholds, and functional sensibility of the hand. A normal result with the SWM and/or M2PD had a good predictive value for normal functional sensibility. Sensitivity was reasonable against object recognition and texture discrimination as reference tests (80-90% and 88-93%), but poor against size discrimination and dot detection (50-75% and 43-65%). Specificity was high for most combinations of SWM or M2PD with any of the tests of functional sensibility (85-99%). Above a monofilament threshold of 2 g, the predictive value of an abnormal test was 100% for dot detection and 83-92% for texture discrimination. This indicates that impairment of touch sensibility at this level correlates well with loss of dot detection and texture discrimination in patients with leprous neuropathy. For M2PD the pattern was very similar. Above a threshold of 5 mm, % of affected hands had loss of dot detection and 73-80% had loss of texture discrimination. Conclusions: Monofilament testing and M2PD do not seem suitable as proxy measures of functional sensibility of the hand in leprosy patients. However, a normal threshold with monofilaments and/or M2PD had a good predictive value for normal functional sensibility. Above a monofilament threshold of 2 g and/or a M2PD threshold of 5 mm, most hands had abnormal texture discrimination. INTRODUCTION Impairment of nerve function is considered the most serious complication of leprosy. 1,2,3,4,5,6 Impairment often leads to disability and handicap of the affected person. 3,7,8 These functional aspects of nerve damage in leprosy have received very little attention. 7 Different modalities of nerve function have been studied in leprosy patients. They include nerve conduction 9,10,11,12 and autonomic vasoregulation, 13,14 as well as simple methods that are suitable for field use, like voluntary muscle testing and sensory testing with nylon filaments or ballpen. 12,15,16,17,18 It has been claimed that touch thresholds determined with SWM correlate well with the functional sensibility of the hand and with what the patient is (still) able to do with his/her hand. 19,20 This is likely to be the outcome measure of treatment that the patient will be most interested 1

3 in. However, the (published) evidence is not very strong and the data supporting this claim originate from a non-leprosy patient population. 21 Hand function depends on many variables including motor function, tactile sensibility, proprioception and cognition. In the field of leprosy touch/pressure sensibility is often the only sensory modality that is evaluated. 22,23 We were interested to investigate to what extent functional sensibility (see below) was affected in leprosy patients and to examine the relationship between touch sensibility and functional sensibility. Therefore, we conducted a study to compare the results of Semmes- Weinstein monofilament testing (SWM) and moving 2-point discrimination (M2PD) with four tests of functional sensibility adapted for use in Nepal: object recognition, size discrimination, dot detection and texture discrimination. The aims of the study were to, 1) examine the validity of the SWM and M2PD as proxy measures for functional sensibility in leprosy patients, and 2) investigate whether the SWM and M2PD can be used as screening tests for the presence or absence of impairment of functional sensibility. METHODS, CONCEPTS AND DEFINITIONS Selection of patients No particular selection criteria or randomization methods were used, as the objective was test-comparison within the same patient. Those who took part in the study included patients admitted to Green Pastures Hospital (GPH) for treatment of leprosy reactions and/or nerve function impairment (NFI) between March and May 1993, as well as patients attending outpatient clinics at GPH during that period. All patients had an established diagnosis of leprosy. Impairment of touch sensibility ranged from no detectable impairment to complete anaesthesia of both hands. In patients with missing digits or severe deformities on one hand, only the remaining hand was tested. Diagnosis Diagnosis of leprosy was essentially clinical, supported by skin smear examination. The diagnosis leprosy was based on finding at least one of three cardinal signs: anaesthetic skin lesions, one or more definitely enlarged peripheral nerve trunks or acid fast bacilli in a slit skin smear. 24 Further details of diagnosis and classification, and laboratory examinations and histology have been described in detail in a previous paper. 25 Functional sensibility Stereognosis (stereos = solid, gnosis = knowing) and functional sensibility have very similar meanings. Butterworths Medical Dictionary defined stereognosis as the ability to recognize the shape and characteristics of an object by means of touch. 26 Collins Dictionary of Medicine qualified characteristics further as shape, size and texture of an object. 27 However, the term is used differently in neurological and in hand surgery literature. In the latter, stereognosis is used to denote the sensory function of the hand. In neurology the term is used for a central function that integrates a variety of sensory 2

4 impulses into a particular pattern that allows someone to recognize an object. 28,29 Thus, in the neurological sense, it is possible to have intact peripheral sensibility and still have a-stereognosis. This is, in fact, the definition of astereognosis. 29 Because of the (potential) confusion concerning the meaning of stereognosis, a different term is preferable. Moberg used the term tactile gnosis. 30 In this paper we use the term functional sensibility to denote the particular sensory perception that enables us to recognize and discriminate by touch. We defined functional sensibility as: the ability of a person to explore and discriminate between, and/or identify objects, including their shape, size and texture, by means of touch. It is very similar to what Gibson described as active touch. 31 At our hospital, four tests of functional sensibility were adapted from existing tests, for use with mostly rural patients. Object recognition This test was first introduced by Moberg in 1958 as a test of tactile gnosis. 28 In the original test the patient was blindfolded and then asked to pick up each of 10 objects one by one, identify them and put them in a container. This process was timed and the result expressed in seconds. Dellon modified the test to timed identification, to limit the effect of concurrent motor function impairment. 32 We introduced a further modification in leaving out the time element, because we considered the ability to recognize objects more important than the speed with which this can be done. Ten objects with a similar texture (safety pin, small padlock, hairpin, nail, button, marble, coin, ring, bottle-lid and key) that were considered common objects in rural Nepal were chosen. First the test was explained and the patient was asked to name the objects by sight, to ensure that all the objects were familiar. Then the patient was asked to close his eyes and was given each of the 10 objects once in random order. The score was the number of items correctly identified out of 10. The normal level was found to be 9/ Size discrimination This test consisted of two identical series of four wooden cubes with sides measuring 1 inch, 1½ inch, 2 inch and 2½ inch. After the test was explained, the patient was asked to close his eyes and he was given one of the cubes (in random order). He then had to identify the size-matched cube out of the duplicate series. To prevent identification by the different weights of the cubes, the patient was not allowed to lift them during the test. The number of cubes correctly identified out of four decided the score of this test. The normal level was 4 out of Dot detection Originally we intended to devise a test for graphaesthesia. Butterworths Medical Dictionary defines graphaesthesia as the ability to recognize letters or figures traced on the skin by blunt pressure. 26 It is the sensory modality which is used by the blind for braille reading. Graphaesthesia is usually tested by tracing letters or figures on the limbs, which must then be recognized by the person tested. Because many of our Nepali patients are unfamiliar with letters or geometric shapes, we felt that the 3

5 traditional way of testing graphaesthesia would not be appropriate. Detection of a small dot on a smooth surface was described by Johansson & LaMotte 34 and proposed by LaMotte & Whitehouse as a means of clinically testing impaired tactile sensitivity in the glabrous skin of the hand. 35 Our testing instrument consisted of an approximately 0.5 mm high and 0.5 mm wide braille-like dot, punched in one corner of a 2x2 cm smooth aluminium square. The square was glued onto a small wooden cube to make it easier to hold. It was presented to the patient, who, with his eyes closed, was asked to identify in which of the four corners the dot was located, using either his index or little finger. The examiner held the testing device between two fingers throughout the test. The number of correct responses out of 5 trials gave the score for this test. Based on our results with healthy volunteers, we used a normal level of 4/5 for the index finger and 3/5 for the little finger. 33 Because the element of recognition was missing from our test, it was not equivalent to a test of graphaesthesia. Therefore, we will call this dot detection rather than graphaesthesia. Texture discrimination This was defined as the ability to discriminate between surfaces of different texture by touch. Humans are able to accurately discriminate small differences in spacing between rows and columns of small dots on a smooth surface with their fingertips. 36 Greenspan & LaMotte have suggested that standardized textures in the form of grating patterns could be used to clinically detect impaired sensibility. 37 We used a test in which five common textures (smooth vinyl, 2 grades of sand paper and 2 kinds of textile with different textures) were glued on wooden square blocks of 10x10 cm. There was little difference between the hardness of each surface. After explanation of the test, the patient was given one surface to feel with the volar pad of either the index or little finger. With eyes closed he had to match this with the same texture block in a duplicate series. This was repeated five times in random order. The score consisted of the number of correct matches out of 5 trials; normal being 3/5 for both index and little fingers. 33 Semmes-Weinstein Monofilament Test (SWM) Touch/pressure sensibility was tested with a standard set of 5 Semmes-Weinstein monofilaments as described by Bell-Krotoski. 38 A score of 5 was given when the thinnest monofilament in the test series was felt; a score of zero if even the thickest filament was not felt. These filaments (Semmes-Weinstein log numbers 2.83, 3.61, 4.31, 4.56 and 6.65) are equal to approximate application forces of 70 mg, 200 mg, 2 g, 4 g and about 280 g when applied with enough force to bend the filament. 39 The normal threshold for the SWM was found to be 200 mg. 40 The following sites were tested: Median nerve: the volar skin of the distal phalanx of the index finger, over the 2nd metacarpophalangeal joint and on the distal phalanx of the thumb. Ulnar nerve: the volar skin of the distal phalanx of the little finger, over the 5th metacarpophalangeal joint and proximally on the hypothenar eminence. 4

6 Moving 2-Point Discrimination (M2PD) Moving touch sensibility of the median and ulnar nerves was assessed with M2PD as recommended by Dellon. 41 This test can be done with a paperclip, but for this study we used a Disk-Criminator TM, * a plastic disk on which pairs of metal prongs are mounted with different inter-prong distances. Details of the testing technique were given in a previous paper. 18 The normal threshold in healthy volunteers was 4 mm for both index and little fingers. 40 Only index and little fingers were tested, at the same sites as for the SWM. Diagnosis of nerve function impairment Impairment was defined as a score below the level or threshold that we chose to represent normal sensibility based on our normative study. 33 Nerve function assessment SWM thresholds and M2PD were tested by trained physiotechnicians, while the tests of functional sensibility were performed by one of the authors (MK, MvL or IBK). Outcome measures 1) Sensitivity and specificity, predictive value of positive and negative monofilament and M2PD test results, 2) Correlation between SWM/M2PD and functional sensibility, 3) Cutoff thresholds of SWM and M2PD (if any), beyond which functional sensibility is lost. Statistical Methods The difference between proportions was tested using the Standard Normal Deviate (SND) for unpaired sample proportions and McNemar s paired Chi-square test for paired sample proportions. 42 The validity of the SWM and M2PD was examined using the tests of functional sensibility as reference tests. It was realised that these tests were not gold standards, but we considered them as representative of functional sensibility. For these calculations the data were recoded as binary variables, using the normal level as cutoff for positive/negative. Because the data were often not normally distributed and because most were graded on ordinal, non-interval scales, we examined the strength of the association between two measures with a non-parametric method, the Spearman rank correlation coefficient. A p-value of less than 5% was used as the level of statistical significance. The 95% confidence interval is given for proportion and correlation coefficients. Analysis was done, using Epi Info software, vs and SPSS for Windows, vs. 6. * Available through P.O. Box 13692, Baltimore, Maryland, 21210, USA 5

7 RESULTS Patients Ninety-eight patients were included in the study. The mean age was 40 years (range 15-63); seventy-nine percent were male and 95% were manual workers. The classification distribution was as follows: tuberculoid 2, borderline tuberculoid 29, borderline 6, borderline lepromatous 32, lepromatous 20, pure neuritic 4 and not classified 5. There was no significant difference in results between right and left hands for any of the tests. We assumed that the presence or absence of an association between functional sensibility and touch sensibility would be independent for the right and left hand of each patient and, therefore, pooled them as one sample of 196 hands. Comparison of nerve function impairment between median and ulnar nerves Table 1 shows the prevalence of nerve function impairment (NFI) by test and by site in our sample. The ulnar nerve was affected much more frequently than the median nerve in all of the tests. There were considerable differences in between-sites monofilament results. The thumb was affected only in 3.1% of the hands, while the index finger was abnormal more than twice as often (7.7%; difference 4.6%, p = 0.007, McNemar s test). For the ulnar nerve the least affected site was the hypothenar (17%), compared with 29% for the little finger (difference 12%, p < , McNemar s test). Cut-offs for predicting functional status For predicting abnormal size discrimination and object recognition no clinically meaningful cutoff thresholds could be defined, except perhaps for the monofilament score of the thumb vs object recognition: among six hands that had an abnormal monofilament threshold (> 200 mg), only one had normal object recognition. For predicting dot detection and texture discrimination outcome the pattern was fairly consistent. Above a monofilament threshold of 2 g the number of hands with abnormal dot detection was 6/6 for the index finger and 38/38 for the little finger. At the same threshold, the number of hands with abnormal texture discrimination was 5/6 for the index finger and 34/37 for the little finger. The M2PD results showed a similar pattern. Above a threshold of 5 mm, 10/10 index fingers and 38/48 little fingers had abnormal dot detection. The same threshold was less good in predicting abnormal texture discrimination: 8/10 for the index finger and 35/48 for the little fingers. Functional sensibility and the median nerve The agreement between the monofilament thresholds at different sites, M2PD and each test of functional sensibility is shown in Table 2. The SWM median combined score (3 sites) gave the highest sensitivity for object recognition (80%), with a specificity of 91%. Spearman s correlation coefficient for this combination was 0.44 ( ). Spearman s correlation coefficient for the SWM score and dot detection of the index finger was 0.5 ( ). For texture discrimination and M2PD of the index finger, the correlation coefficient was 0.25 ( ). 6

8 Functional sensibility and the ulnar nerve Table 3 shows the agreement between the tests for the ulnar nerve. Surprisingly, the monofilament result of the little finger agreed well with the object recognition score. Spearman s correlation coefficient was 0.35 ( ). The correlation coefficients for the combination ulnar combined and dot detection, and little finger and dot detection were both high, 0.77 ( ) vs 0.73 ( ). There was good agreement between the ulnar SWM and M2PD scores and texture discrimination. The combination of the SWM ulnar score and texture discrimination had a correlation coefficient of 0.69 ( ). DISCUSSION Some investigators have claimed a close relationship between hand function and Semmes-Weinstein monofilament testing, 21,44 while others have said that the two do not correlate well. 32,45 Moberg and Dellon reported that the tests that correlate best with functional sensibility are static and M2PD. 28,32,46 In a previous study, we found a good correlation between results of SWM and M2PD in sensory assessment of leprosy patients. 18 The present study shows that, although there is no strong correlation between touch sensibility scores and the results of the functional sensibility tests used, when using the right cutoff, both the SWM and the M2PD test may be used as a predictor of functional sensibility. Object recognition This test combines size discrimination, graphaesthesia and texture discrimination into one and could, therefore, be seen as a standalone test for functional sensibility. It also represents hand function more comprehensively than any of the other tests, as it involves motor function to some extent. Citron & Taylor considered a timed object recognition test, which they called the functional recognition test, to be a reliable and reproducible test of sensory function in patients with peripheral nerve injury. 47 Only 10/196 hands showed impaired object recognition. The more complex skill of object recognition was not affected earlier than a more basic one, like texture discrimination or touch. Loss of object recognition may be related to impairment of proprioception, which was uncommon in our patient population. 18 Surprisingly, sensitivity values for object recognition were higher for the ulnar than for the median nerve, while manipulation of the objects was mainly done between thumb and index finger. However, it should be noted that these values are not very reliable, because of the small number of hands scoring abnormally on this test. The predictive value of a normal monofilament and moving 2-point discrimination test was high, suggesting that a patient with normal touch thresholds is also likely to have normal object recognition. Dellon & Kallman used a timed-object recognition test as a measure of functional sensibility in 18 patients with nerve lesions due to injuries or compression. 46 M2PD correlated much better with the number of objects identified than pressure 7

9 sensibility measured with SWM (Pearson s r = 0.87 vs 0.45). Novak et al. evaluated an object recognition test similar to ours, with eight objects. 45 Inter-tester reliability (between 2 testers) was very high (intraclass correlation coefficient (ICC) 0.99). Correlation between the results of this test and results of M2PD was close, with a Spearman correlation coefficient of Correlation with monofilament thresholds was less good (correlation coefficient -0.69). In the present study correlation was better between functional sensibility and the SWM results than between the former and M2PD. Thus it seems that the relationship between impairment of touch sensibility and object recognition varies with the nature of the nerve lesion. Size discrimination A human being can accurately discriminate distances between thumb and fingers. 31 Only very few of our patients had impairment of size discrimination. There was little correlation between the tests of touch sensibility and size discrimination. This suggests either that the sensory mechanisms involved in discriminating sizes are mediated by fibre systems that are not usually affected by leprosy, or that the two modalities are not associated in the first place. Dot detection Johansson and LaMotte showed that humans with intact sensibility can detect raised elements of only about 3 microns high and 230 microns diameter on a smooth surface. 34 Movement of the fingertip was essential for detection. Rapidly adapting mechanoreceptors were shown to have the lowest detection thresholds, 35 but when bigger dots were used, other mechanoreceptive afferents were stimulated too. Our dot detection test was very much a supra-threshold test because of the bigger size of the dot. The test result was abnormal in most patients with an abnormal SWM or M2PD threshold, but it gave a wide range of results in patients with normal touch thresholds. This may mean that it was more sensitive to detect sensory impairment, but such a conclusion would need to be confirmed, preferably in a test battery that includes a controlled-stimulus instrument such as a vibrometer. 8

10 In another recent study Novak et al. reported on a similar new measure of fine sensory function, the braille pattern identification test. 48 Instead of one braille dot, as in our test, this test used several patterns in matrices of 3x3. Because this test involved recognition, it can be said to be a test of graphaesthesia. They found good correlation between this new test and 2-point discrimination (static and moving), with an intraclass correlation coefficient (ICC) of around Correlation with vibration thresholds and SWM was slightly less good (ICC around -0.65). Porter reported a good correlation between his test of letter recognition, 2-point discrimination and Moberg s pick-up test. 49 (Moving) 2-point discrimination thus seems to correlate well with tests involving an element of recognition, like object recognition and classical graphaesthesia testing. The fact that we removed the recognition element from our test may, therefore, explain the lack of correlation between M2PD and dot detection. Texture discrimination The ability to discriminate differences in roughness is mediated by the SA I and RA I mechanoreceptors in the skin. 36,50 Thus, given normal cognition and proximal fibre systems, texture discrimination would test the integrity of the thick myelinated afferents. It is therefore not surprising that this test correlated well with monofilament thresholds and moving 2-point discrimination. Both sensitivity and specificity of the latter two were high. A normal touch sensibility test result was predictive of normal texture discrimination. Texture discrimination correlated better with monofilament thresholds than with M2PD (Spearman s correlation coefficients 0.70 vs 0.59). Novak et al. evaluated reliability of texture identification, in which 30 patients were asked to order five cards with different textures from smooth to coarse. 45 Both time and order of response were considered. The agreement between two testers was good (Intraclass correlation coefficient 0.77). They also examined correlation between texture identification, monofilament testing and M2PD. The Spearman correlation coefficient for the combination texture identification vs monofilament testing was or -0.44, depending on the examiner. For the combination texture identification vs M2PD the coefficients were and Thresholds for predicting abnormal functional sensibility No clinically meaningful thresholds could be identified to predict abnormal object recognition and size discrimination. This was partly due to the very high number of patients scoring normally on these tests. For dot detection and texture discrimination the pattern was consistent. Above a monofilament threshold of 2 g, the predictive value of an abnormal test was 100% for dot detection and 83-92% for texture discrimination. This monofilament threshold is said to correspond with protective sensation in the hand. 21 It may be that loss of protective sensation is associated with impairment of dot detection and texture discrimination in patients with leprous neuropathy. The pattern was very similar for M2PD. Above a threshold of 5 mm most affected hands had impairment of dot detection and texture discrimination. The corresponding predictive 9

11 value of an abnormal test at this threshold (73-80%) was slightly less good than that of the monofilament test. It should be noted that the predictive value of a test is dependent on the prevalence of the condition that is measured. 51 In situations where the prevalence of neural impairment is lower than in the present study, proportionally more patients will test false positive and, therefore, the predictive value of a positive (abnormal) test result will be lower. The above results suggest that in leprous neuropathy there is only a moderate correlation between monofilament or moving 2-point test results and functional sensibility. There may be three explanations for this finding: 1) Touch/pressure sensibility and M2PD may not be suitable as proxy tests for functional sensibility of the hand. Actively exploring fingers use all available receptors (including mechanoreceptors in joints and muscles) to generate stimuli that give information about the object that is touched. Unless all afferent fibres are affected homogeneously, which is not true in leprosy, 52 a close correlation between touch thresholds and functional sensibility cannot be expected. 2) The four tests that were used in this study may not have been valid and reliable enough as measures of functional sensibility. No reference test for functional sensibility was available against which our tests could be validated. Tests of hand function that are commonly used in rehabilitation or occupational therapy assessments in developed countries include the Jebsen hand function test, the Williams board test, and Moberg s precision sensory grips. 53,54,55 None of these tests was suitable in their present form to be used as a reference test in the Nepali context. 3) The range of impairment in our sample could be considered inadequate. With a prevalence of median and ulnar sensory impairment of 12% and 31% (SWM), most patients were expected to have normal functional sensibility. Only 1.5% of median nerves and 3.6% of ulnar nerves in our patient sample had complete loss of touch sensibility. Therefore, it could be argued that there was an insufficient number of cases with severe nerve damage to fully examine the relationship between SWM, M2PD and the tests of functional sensibility. In a non-referral centre situation, however, one usually deals with patient populations that have a lower prevalence of neuropathy than ours and so our findings may well represent the operational relationship between the tests. In our experience, the consequences of impairment of sensibility are still inadequately appreciated by many leprosy workers. Our understanding of the sensory neuropathy in leprosy and its epidemiology and treatment is still only partial. Much further work is needed on areas including functional assessment of hands and feet, and prevention and treatment of neuropathy. But however limited our knowledge or resources may be, much harm can be prevented if physicians and health workers treating leprosy patients would only use available methods of nerve function assessment regularly and treat patients when necessary. 10

12 CONCLUSIONS 1. Semmes-Weinstein monofilament testing and moving 2-point discrimination do not seem suitable as proxy measures of functional sensibility of the hand in leprosy patients. 2. A normal threshold with monofilaments and/or M2PD has a good predictive value for normal functional sensibility. 3. Above a monofilament threshold of 2 g and/or a M2PD threshold of 5 mm, most hands had abnormal texture discrimination. These results support the validity of the SWM and M2PD as screening tests for impairment of tactile sensibility in leprosy patients. ACKNOWLEDGEMENTS The authors wish to thank Dr. Yolanda van der Graaf and Prof. Dr. Frans G.I. Jennekens for constructive criticism on a previous draft of the manuscript. We are indebted to the staff of the physiotherapy department at Green Pastures Hospital who spend much of their time performing detailed functional assessments, on which this study was based. The work at Green Pastures Hospital is dedicated to the service and glory of God. 11

13 Median nerve Ulnar nerve Test/Site FI a (%) 95% CI b Site FI a (%) 95% CI b SWM c thumb 3.1 ( ) little finger 29 (23-35) mcp2 d 11 (5.8-14) mcp5 e 22 (16-28) index finger 7.7 (4-11) hypothenar 17 (12-22) median combined f 12 (7-17) ulnar combined f 31 (25-37) M2PD g index finger 7.7 (4-11) little finger 30 (24-36) OBJECT RECOGNITION 5.1 (2-8.2) SIZE DISCRIMINATION 2.1 ( ) DOT DETECTION index finger 19 (14-24) little finger 45 (38-52) TEXTURE DISCRIMINATION index finger 4.1 ( ) little finger 22 (16-28) a function impairment, b 95% confidence interval, c Semmes-Weinstein monofilament test, d 2nd metacarpophalangeal joint, e 5th metacarpophalangeal joint, f combined score for the 3 sites, which was called abnormal (=impairment) if any of the 3 sites was anaesthetic for the 200mg filament, g moving 2-point discrimination Screening test Reference test cut-off level Object recognition Predictive value (%) abnormal (positive) normal (negative) sensitivity (%) specificity (%) SWM a thumb 200mg 5/6 d 97 5/10 d 99 SWM index finger 200mg 6/ /10 95 SWM median b 200mg 8/ /10 91 M2PD c index finger 4mm 5/ /10 94 Size discrimination SWM thumb 200mg 2/5 99 2/4 98 SWM index finger 200mg 2/ /4 94 SWM median 200mg 2/ /4 89 M2PD index finger 4mm 1/ /4 92 Dot detection SWM index finger 200mg 11/ /37 97 SWM median 200mg 16/ /37 95 M2PD index finger 4mm 11/ /37 96 Texture discrimination SWM index finger 200mg 6/ /8 96 SWM median 200mg 6/ /8 91 M2PD index finger 4mm 7/ /8 95 a Semmes-Weinstein monofilament test, b 3 sites combined (thumb, 2nd metacarpophalangeal joint and index finger), c moving 2-point discrimination test, d When the denominator is small (<40), the fraction is given instead of the percentage.

14 Screening test Reference test cut-off level Object recognition Predictive value (%) abnormal (positive) normal (negative) sensitivity (%) specificity (%) SWM a little finger 200mg /10 d 75 SWM ulnar b 200mg /10 73 M2PD c little finger 4mm /10 72 Size discrimination SWM little finger 200mg /4 72 SWM ulnar 200mg /4 70 M2PD little finger 4mm /4 70 Dot detection SWM little finger 200mg SWM ulnar 200mg M2PD little finger 4mm Texture discrimination SWM little finger 200mg SWM ulnar 200mg M2PD little finger 4mm a Semmes-Weinstein monofilament test, b 3 sites combined (little finger, 5th metacarpophalangeal joint and hypothenar, c moving 2-point discrimination test, d When the denominator is small (<40), the fraction is given instead of the percentage.

15 REFERENCES 1. Ridley DS, Job CK. The pathology of leprosy. In: Hastings RC. (ed) Leprosy. Churchill Livingstone, Edinburgh, Job CK. Nerve damage in leprosy. Int J Lepr 1989; 57: Srinivasan H. Prevention of disabilities in patients with leprosy. A practical guide. World Health Organisation, Geneva, Watson JM. Essential action to minimise disability in leprosy patients. The Leprosy Mission International. Stanley L. Hunt (Printers) Ltd, Becx-Bleumink M, Berhe D, 'T Mannetje W. The management of nerve damage in the leprosy control services. Lepr Rev 1990; 61: Rose P, Waters MFR. Reversal Reactions in leprosy and their management. Lepr Rev 1991; 62: Brandsma JW, Heerkens YF, Lakerveld-Heyl K, Mischner-van Ravensberg CD. The international classification of impairments, disabilities and handicaps in leprosy-control projects. Lepr Rev 1992; 63: Coleman WC, Madrigal DR. The patient with sensory loss. Int J Lepr 1985; 53(2): Magora A, Sheskin J, Sagher F, Gonen B. The condition of the peripheral nerve in leprosy under various forms of treatment. Int J Lepr 1970; 38: Verghese M, Ittimani KV, Satyanaran KR, Mathai R, Bhakthaviriam C. A study of the conduction velocity of the motor fibres of the ulnar and median nerves in leprosy. Int J Lepr 1970; 38: McLeod JG, Hargrave JC, Walsh JC, Booth GC, Gye RS, Barron A. Nerve conduction studies in leprosy. Int J Lepr 1975; 43: Naafs B, Dagne T. Sensory testing: a sensitive method in the follow-up of nerve involvement. Int J Lepr 1977; 45: Abbot NC, Beck JS, Samson PD, Butlin CR, Brown RA, Forster A, Grange JM, Cree IA. Impairment of fingertip vasomotor reflexes in leprosy patients and apparently healthy contacts. Int J Lepr 1991; 59: Abbot NC, Beck JS, Samson PD, Butlin CR, Bennett PJ, Grange JM. Cold fingers in leprosy. Int J Lepr 1992; 60: Birke JA, Sims DS. Plantar sensory threshold in the ulcerative foot. Lepr Rev 1986; 57: Watson JM. Disability control in a leprosy control programme. Lepr Rev 1989; 60: Palande DD, Bowden REM. Early detection of damage to nerves in leprosy. Lepr Rev 1992; 63: van Brakel WH, Shute J, Dixon JA, Arzet H. Evaluation of sensibility in leprosy - comparison of various clinical methods. Lepr Rev 1994; 65 (in press). 19. Bell-Krotoski JA. Hand screen for early detection and monitoring of peripheral neuropathy part II. The Star 1992; 51(3): Breger D. Correlating Semmes-Weinstein monofilament mappings with sensory nerve

16 conduction parameters in Hansen s disease patients: an update. J Hand Ther 1987; Oct- Dec: Bell-Krotoski JA. Light touch-deep pressure testing using Semmes-Weinstein monofilaments. In: Rehabilitation of the hand, 3rd ed. Hunter et al.(eds) C.V. Mosby Co., 1989; Prevention of disability. Guidelines for leprosy control programmes. ILEP, London, Jennekens FGI, Jennekens-Schinkel A. Neurological examination of patients suffering from leprosy: is it worthwhile? Lepr Rev 1992; 63: Pfaltzgraff RE, Bryceson A. Clinical leprosy. In: Hastings RC. (ed), Leprosy. Edinburgh: Churchill Livingstone, 1985: p van Brakel WH, de Soldenhoff R, McDougall AC. The allocation of leprosy patients into paucibacillary and multibacillary groups for multidrug therapy, taking into account the number of body areas affected by skin, or skin and nerve lesions. Lepr Rev 1992; 63: Butterworths Medical Dictionary. Critchley M (chief ed.), Butterworths & Co (Publ), London, Collins Dictionary of Medicine. Youngson RM. HarperCollins (Publ), Glasgow, Moberg E. Objective methods of determining functional value of sensibility in the hand. J Bone Jt Surg (B) 1958; 40: Luria AR. Higher cortical functions in man. Basic Books, Inc, Publishers, Cambridge (Mass.), Moberg E. Criticism and study of methods for examining sensibility in the hand. Neurology 1962; 12: Gibson JJ. Observations on active touch. Psychol Rev 1962; 69: Dellon AL. Evaluation of sensibility and re-education of sensation in the hand. Baltimore: John D. Lucas Printing Co, van Brakel. Peripheral neuropathy in leprosy - the continuing challenge. Doctorate thesis, University of Utrecht, Johansson RS, LaMotte RH. Tactile detection thresholds for a single asperity on an otherwise smooth surface. Somatosens Res 1983; 1: LaMotte RH, Whitehouse J. Tactile detection of a dot on a smooth surface: peripheral neural events. J Neurophys 1986; 56: Connor CE, Johnson KO. Neural coding of tactile texture: comparison of spatial and temporal mechanisms for roughness perception. J Neurosc 1992; 12: Greenspan JD, LaMotte RH. Cutaneous mechanoreceptors of the hand: experimental studies and their implications for clinical testing of tactile sensation. J Hand Ther 1993; Apr-June: Bell-Krotoski JA. "Pocket" filaments and specifications for the Semmes-Weinstein Monofilaments. J Hand Ther, 1990; Jan-Mar: Bell-Krotoski JA, Tomancik E. The repeatability of testing with SemmesWeinstein Monofilaments. J Hand Surg, 1987; 12A:

17 40. Kets M, van Leerdam M, van Brakel WH, Khawas IB, Gurung KS. Reference values touch sensibility thresholds in healthy Nepalese volunteers. Lepr Rev 1996; 67: 41. Dellon AL. The moving two-point discrimination test: clinical evaluation of the quickly adapting fiber/receptor system. J Hand Surg, 1978; 3: Armitage P, Berry G. Statistical Methods in Medical Research. 2nd ed. Oxford: Blackwell Scientific Publications, 1987: Dean AG, Dean JA, Dicker RC. Epi Info, Version 5: a word processing, database, and statistics program for epidemiology on microcomputers. USD, Inc., Stone Mountain, Georgia, Bell-Krotoski JA, Weinstein S, Weinstein C. Testing sensibility, including touch-pressure, two-point discrimination, point localization and vibration. J Hand Ther 1993; Apr-June: Novak CB, Mackinnon SE, Williams JI, Kelly L. Establishment of reliability in the evaluation of hand sensibility. Plast Recon Surg 1993; 92(2): Dellon AL, Kallman CH. Evaluation of functional sensation in the hand. J Hand Surg (A), 1983; 8: Citron N, Taylor J. Tendon transfer in partially anaesthetic hands. J Hand Surg 1987; 12-B: Novak CB, Mackinnon SE, Williams JI, Kelly L. Development of a new measure of new sensory function. Plast Recon Surg 1993; 92(2): Porter RW. New test for finger-tip sensation. Br Med J 1966; 2: Phillips JR, Johansson RS, Johnson KO. Responses of human mechanoreceptive afferents to embossed dot arrays scanned across fingerpad skin. J Neurosc 1992; 12: Hennekens CH, Buring JE. Epidemiology in medicine. Mayrent SL. (ed). Little, Brown and Co., Boston, 1987: pp. 336 ff. 52. Jacobs JM, Shetty VP, Antia NH. Teased fibre studies in leprous neuropathy. J Neurol Sci 1987; 79: Hackel ME, Wolfe GA, Bang SM, Canfield JS. Changes in hand function in the aging adult as determined by the Jebsen test of hand function. Phys Ther 1992; 72: Falconer J, Hughes SL, Naughton BJ, Singer R, Chang RW, Sinacore JM. Self report and performance-based hand function tests as correlates of dependancy in the elderly. J Am Geriatr Soc 1991; 39: Moberg E. Criticism and study of methods for examining sensibility in the hand. Neurology 1962; 12: 8-19.