SEARCH

SEARCH BY CITATION

Keywords:

  • Semmes-Weinstein monofilament;
  • loss of protective sensation;
  • false-positive

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

Aims

The Semmes-Weinstein monofilament is the most widely used test to diagnose the loss of protective sensation. The commonly used protocol of the International Consensus on the Diabetic Foot includes a ‘sham’ application that allows for false-positive answers. We sought to study the heretofore unexamined significance of false-positive answers.

Methods

Forty-five patients with diabetes and a history of pedal ulceration (Group I) and 81 patients with diabetes but no history of ulceration (Group II) were studied. The three original sites of the International Consensus on the Diabetic Foot at the hallux, 1st metatarsal and 5th metatarsal areas were used. At each location, the test was performed three times: 2 actual and 1 “sham” applications. Scores were graded from 0 to 3 based upon correct responses. Determination of loss of protective sensation was performed with and without calculating a false-positive answer as a minus 1 score.

Results

False-positive responses were found in a significant percentage of patients with and without history of ulceration. Introducing false-positive results as minus 1 into the test outcome significantly increased the number of patients diagnosed with loss of protective sensation in both groups.

Conclusions

False-positive answers can significantly affect Semmes-Weinstein monofilament test results and the diagnosis of LOPS. A model that accounts for false-positive answers is offered. Copyright © 2013 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

Diabetic polyneuropathy with loss of protective sensation (LOPS) is a key risk factor for neuropathic ulceration [1-5]. The Semmes-Weinstein monofilament (SWF) 10 g is the most widely used and accepted test to determine LOPS [3]. Although a number of prospective studies have reported a correlation between poor scores on the monofilament test and subsequent ulceration [2, 6-10], there is no standard protocol. Therefore, methods as well as interpretation of the results vary greatly [3-5, 11]. One commonly used protocol, the International Consensus on the Diabetic Foot (ICDF) [5], recommends testing three sites: hallux, metatarsal head 1 and metatarsal head 5.

According to this protocol, two applications and a ‘sham’ application are performed at each site and a score of ≤1 at any single site defines the patient as having LOPS. However, the protocol does not attribute special significance to false-positive answers. Therefore, a patient can inappropriately pass the test in the ICDF protocol by automatically answering ‘yes’ at each application. Furthermore, any score ≤1 at any single site automatically defines a patient as having LOPS even though all answers from all other sites in both feet may be normal. This inherent weakness creates an unacceptable dependency of the SWF test on a single site and may result in a wrong diagnosis of LOPS. To date, there has been no investigation of the significance of false-positive answers on test outcome and the resulting diagnosis of LOPS.

Our study was designed to determine for the first time the significance of false-positive scores in the SWF test and to provide a modified scoring system accounting for the false-positive responses.

Research design and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

Patients

Enrolled patients presented to the Diabetic Foot Clinic of Assaf Harofeh Medical Center between September, 2006 and September 2008. Forty-five diabetic patients with known history of pedal ulceration (Group I) and 81 diabetic patients without history of ulceration (Group II) were studied (Table 1). All patients in Group I had been treated for pedal ulceration in our clinic; no patient in this group had an active ulcer at the time of testing; 23 of the patients had a history of multiple ulceration sites with 14 having a history of recurrence. Patients who had undergone amputation were excluded. The medical history, including presence of retinopathy and nephropathy, was documented from a letter of referral by the primary-care physician. Peripheral neuropathy was determined by the presence of neurological symptoms such as numbness and paresthesias. Haematology and blood chemistry tests that were done within 3 months of the date of the examination were available for all patients. All subjects who met the inclusion criteria consented to participate in this study. The ethics committee approved the study protocol.

Table 1. Demographic and clinical characteristics of the study patients
Group (# of patients)Age (years)Male/femaleBMIT2DMDisease durationHba1c (%)History of complications: neuropathy or retinopathy or nephropathyPVD Presence of pulses ± DopplerDistribution of ulcers Lt/RtLocation of ulcers
1 (45)63.7 ± 1.430/1530 ± 0.84223 ± 5.38 ± 0.2Having at least one complication: 40/45 (89%).7/45 (16%)52/47Hallux 31%, toes 26%, metatarsals 27%, others 15%
Retinopathy: 31/45 (69%)
Nephropathy: 21/45 (47%)
Neuropathy: 26/45: (58%)
2 (81)59.9 ± 1.240/4129.3 ± 0.56915.8 ± 17.75 ± 0.2Having at least one complication 63/81 (78%)6/81 (7.5%)
Retinopathy: 46/81: (57%)
Nephropathy: 36/81: (44%)
Neuropathy: 43/81: (53%)

Monofilament testing

Patients sat in a custom treatment chair with legs elevated. A screen was placed at the level of the knees so that patients could not see their lower legs and feet. The SWF 10 g (USA government LEAP program) test was explained and demonstrated first on the patient's hand in front of the screen and then on the foot behind the screen with repetition as necessary in order to ensure the patient understood the test. The standard protocol of the ICDF was used [5]. The three original sites of the ICDF at the hallux, 1st metatarsal and 5th metatarsal areas were used. No SWF was used on more than one patient in any given week. At each location, the test was performed three times: 2 actual and 1 “sham” applications. The sequence of the “sham” and actual applications was randomly altered. Patients and controls were asked for a yes/no response after each application. Scores were graded from 0 to 3 based upon correct responses, and false-positive answers were recorded. A score of ≤1 in any of the sites indicated LOPS.

Statistical analysis

The data were analysed using BMDP [12]. Student's t-test was used for comparison between groups. A p value <0.05 was considered significant. Using logistic regression with the receiver operating characteristic plot, we were able to identify the best cut-off point, from which we were able to present two-by-two tables, thus enabling us to derive the sensitivity and specificity.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

Standard Semmes-Weinstein test – significance of false-positive answers

In order to determine the clinical significance of false-positive answers, we initially calculated the percentage of false-positive answers. At each site, up to 30% (range 24% to 29%) of patients in group 1 responded with a false-positive answer. This was true for both left and right feet. Percentage of false-positive answers was lower but significant in group 2 patients ranging from 3.7% to 11% with a mean of 8%. When test results were recalculated taking into account a false-positive answer as minus 1, a significant increase in the number of patients diagnosed with LOPS; 89% of patients in group 1 and 44% of patients from group 2 were defined as having LOPS as opposed to 75% and 37%, respectively, when a false-positive answer was not taken into account (p = 0.03). These data demonstrate that false-positive answers on the SWF test are common in diabetic patients with and without history of ulceration and can be a significant factor in the evaluation of results.

Effect of modified scoring system on Semmes-Weinstein monofilament test results

In the current SWF protocol, any score ≤1 in any single site automatically defines the patient as having LOPS even though all answers from all other sites in both feet may be normal. In order to overcome this undesirable dependency on single sites, a total score was calculated for all three sites for both feet, and its association with LOPS was evaluated in all patients. A cut-off of 11 correct responses out of 18 resulted in sensitivity of 82% and specificity of 76%, respectively. Calculating a false-positive response as a minus1 improved the sensitivity and specificity of the total scoring system to 87% and 79%, respectively.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

The SWF test is commonly accepted as a leading diagnostic tool for the prediction of neuropathic ulceration in diabetic patients. Nevertheless, the percentage and significance of false-positive response is not known. It is also not known whether the combined scores of all sites have a different significance as compared with individual site scores in terms of correlation with risk of ulceration. Our study sought for the first time to answer these intriguing and clinically important questions in a population of diabetic patients with and without prior history of ulceration

Our data show that a significant percentage of patients with and without history of ulceration responded with false-positive answers. When we integrated a false answer into the scoring as a minus 1, we found a significant increase in the number of patients diagnosed with LOPS in both groups. This finding underscores that the current protocol of the SWF testing contains an inherent weakness because it permits a significant number of patients to have a normal score by simply answering positively to all applications. In order to overcome this weakness, we propose a new scoring system using a total score of the original three sites of the standard ICDF protocol on both feet rather than rely on single sites as well as integrating a false-positive answer as a minus 1. The total-score system eliminates the dependency on a single site to diagnose LOPS. At the same time, it increases the relative weight of false-positive answers without over estimating them and truly fulfils the purpose of the sham application. The maximum total score in this proposed system is 18 based on three correct answers in each of the three sites of each foot. Our data demonstrate that any score below a cut-off of 10 out of 18 indicates a high risk of ulceration, namely LOPS, with a sensitivity of 87% and a specificity of 79%. A similar, although not identical, system was proposed by Olmos et al. with a maximum score of six points based upon one application, without any sham applications, to three sites on both feet. In their study, a cut-off point of 5/6 yielded a sensitivity of 85.7% [13].

Our study suffers from several weaknesses, most importantly, the validity of our proposed modified scoring system. Only a ‘head to head’ prospective study comparing our suggested scoring method and the current method of scoring can tell which method has a better sensitivity and/or specificity in predicting future ulceration. To the best of our knowledge, no such study was performed to evaluate any of the scoring methods for the SWF test, including the one currently recommended by the ICDF. In addition, we did not assess peripheral neuropathy by electromyography/nerve conduction velocity or vibratory studies. These data might have provided us with further insight regarding the relative role of peripheral diabetic neuropathy and the effect of other possible factors such as radiculopathy and compression neuropathy on SWF results.

In conclusion, this retrospective study demonstrates that the current ICDF protocol for SWF testing does not fully account for false-positive responses. Our proposed model, based on total score of multiple sites rather than single site analysis, overcomes this weakness, is practical, and provides acceptable sensitivity and specificity for risk of ulceration. Future prospective studies are needed to determine the role of this newly proposed model in evaluating the risk of ulceration in diabetic patients.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References

R.A.S. researched data and wrote the manuscript. S. K. contributed to the discussion and reviewed/edited the manuscript. Y.R. reviewed/edited the manuscript. A.B. reviewed/edited the manuscript. M.J.R. contributed to the discussion and wrote the manuscript. We thank Adina Slater and Shai Rapoport for technical support.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research design and methods
  5. Results
  6. Conclusions
  7. Acknowledgements
  8. Conflict of interest
  9. References
  • 1
    Boulton AJM, Armstrong DG, Albert SF, et al. Comprehensive foot examination and risk assessment: a report of the task force of the foot care interest group of the American Diabetes Association, with endorsement by the American Association of Clinical Endocrinologists. Diabetes Care 2008; 31: 16791685.
  • 2
    Armstrong DG, Lavery LA, Vela SA, Quebedeaux TL, Fleischli JG. Choosing a practical screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern Med 1998; 158: 289292.
  • 3
    Mayfield JA, Sugarman JR. 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(Suppl. 11): S17S29.
  • 4
    Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA 2005; 293: 217228.
  • 5
    International Consensus on the Diabetic Foot. International Working Group on the Diabetic Foot, 2007.
  • 6
    Boyko EJ, Ahroni JH, Stensel V, Forsberg RC, Davignon DR, Smith DG. A prospective study of risk factors for diabetic foot ulcer: the Seattle Diabetic Foot Study. Diabetes Care 1999; 22: 10361042.
  • 7
    Pham H, Armstrong DG, Harvey C, Harkless LB, Giurini JM, Veves A. Screening techniques to identify people at high risk for diabetic foot ulceration. Diabetes Care 2000; 23: 606611.
  • 8
    Abbott CA, Carrington AL, Ashe H, et al. The North-West Diabetes Foot Care Study: incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort. Diabet Med 2002; 19: 377384.
  • 9
    Rith-Najarian SJ, Stolusky T, Gohdes DM. Identifying diabetic patients at high risk for lower-extremity amputation in a primary health care setting. Diabetes Care 1992; 15: 13861389.
  • 10
    Litzelman DK, Marriott DJ, Vinicor F. Independent physiological predictors of foot lesions in patients with NIDDM. Diabetes Care 1997; 20: 12731278.
  • 11
    Dros J, Wewerinke A, Bindels PJ, van Weert HC. Accuracy of monofilament testing to diagnose peripheral neuropathy: a systematic review. Ann Fam Med 2009; 7: 555558.
  • 12
    BMDP Statistical Software 1993. Chief Editor: W. J. Dixon University of California Los Angeles Press
  • 13
    Olmos PR, Cataland S, O'Dorisio TM, Casey CA, Smead WL, Simon SR. The Semmes-Weinstein monofilament as a potential predictor of foot ulceration in patients with noninsulin-dependent diabetes. Am J Med Sci 1995; 309: 7682.