M. Mcgill et al., Possible sources of discrepancies in the use of the Semmes-Weinstein monofilament - Impact on prevalence of insensate foot and workload requirements, DIABET CARE, 22(4), 1999, pp. 598-602
OBJECTIVE - The purpose of this study was to evaluate the effects of differ
ent testing sites and buckling strengths on the sensitivity and specificity
of using the Semmes-Weinstein monofilament to detect patients with insensa
te foot. The impact on workload required to educate and follow up these hig
h risk individuals was estimated by modeling in our patient population with
a documented status of neuropathy.
RESEARCH DESIGN AND METHODS - Using the 5.07/10-g monofilament, one observe
r tested 132 randomly selected subjects with diabetes at five sites on the
right foot. The sensitivity and specificity of each site and combinations o
f sites in detecting vibration perception threshold >40 was calculated. In
addition, two monofilaments, one with a buckling force of 5 g and the other
with a force of 15 g, were compared by testing 200 randomly selected patie
nts. An estimate of the prevalence of insensate foot and workload was made
by modeling the findings to the 5,270 patients with neuropathy status regis
tered on our computerized database.
RESULTS - Specificity of the 5.07/10-g monofilament to detect insensate foo
t at each of the five sites is high, at similar to 90%, but there is consid
erably more variation and lower sensitivity ranging from 44-71%. Data deriv
ed from the use of different combinations of sites show ed that more string
ent criteria are associated with lower sensitivity but higher specificity.
If the foot is considered insensate when either of sites 3 and 4 (plantar a
spect of the first and fifth metatarsal heads, respectively) cannot feel th
e monofilament, there is reasonable sensitivity and specificity (80-86%, re
spectively). By modeling on our diabetes center population, it can be demon
strated that the choice of different methodologies leads to different concl
usions about the prevalence of severe neuropathy, ranging from 3.4 to 29.3%
.
CONCLUSIONS - Using a combination of sites 3 and 4 for monofilament testing
gives a reasonable compromise for time, sensitivity, and specificity. Mino
r changes in sensitivity and specificity can lead to major changes in the p
revalence of neuropathy with implications for workload.