Sage Journals: Discover world-class research

Abstract

Background:

Neuropathic feet are at very high risk for infection and amputation. The slipping slipper sign (SSS) is elicited by a simple questionnaire test reported to detect the presence of severe diabetic peripheral neuropathy. This test can be administered by non-medical staff. In this study, subjects with and without the SSS were evaluated by nerve conduction studies (NCS) and ultrasound measurements of the right sural nerve diameters as well as with traditional scoring systems for peripheral and autonomic neuropathy.

Objective:

To demonstrate that the Slipping Slipper Sign can be used as an index of severe diabetic peripheral neuropathy

Method:

This was a prospective cross sectional study in which 74 patients with diabetes (38 positive and 36 negative for SSS) underwent ultrasonography and NCS of the right sural nerve by an examiner blinded to SSS status. Findings were evaluated against demography, clinical history, anthropometry as well as traditional clinical and autonomic neuropathic scores.

Results:

Patients without the SSS [median (IQR) = 10.0 years (4.0–20.3)] had a significantly shorter duration of diabetes compared with those with the SSS [median (IQR) = 15.0 years (8.5–25.0)], p = 0.028. The frequencies of retinopathy (36.8% vs 2.8%, p < 0.05) and cerebrovascular accidents (18.4% vs 13.9 %, p < 0.05) were higher among those with SSS compared with those without. Differences in nerve conduction characteristics were markedly significant. The amplitude of the sural sensory nerve action potential (SNAP) was ([median (IQR)] 0 microvolts vs 4.0 microvolts (0.0–10.8) p < 0.002) between those with and without SSS, respectively whilst none of patients with SSS had a recordable SNAP vs 78% without a SSS. Similarly, maximal thickness of the right sural nerve at the ankle 3.0 mm (2.3–3.4) vs 3.5 mm (3.0–3.9), and leg 3.4 mm (2.7–3.8) vs 3.9 mm (3.3–4.2) was reduced, p < 0.01 in patients with the SSS compared with those with a negative SSS.

Conclusion:

The SSS identifies feet with objective neurophysiological and imaging characteristics of severe neuropathy.

Keywords

Slipping Slipper Sign Diabetes Mellitus Diabetic Peripheral Neuropathy Screening test

INTRODUCTION

The ‘diabetic foot’ is a distressing but unfortunately common problem. Diabetic peripheral neuropathy (DPN) is often the underlying factor by predisposing to traumatic ulceration and infection. In such feet, the risk of amputation is high. The reported prevalence of DPN is high in both community-based [1 –3] and hospital-based practices [4] yet it is not uncommon for patients to present late in the course of the disease when complications have already been established [5]. Strategies that could detect DPN earlier or recognize feet most at risk are therefore highly desirable.

However, testing and evaluating DPN are often encumbered by lack of clinical time and resources for formal testing [6]. Most patients the world over are seen and managed by community-based practitioners [7] often without the necessary training, experience or most importantly, clinical time for adequate evaluation of the complexities of the diabetic foot. Slippers [8], defined here as footwear that are casual slip-on footwear on account of being strapless at the ankles, are mainly used indoors in colder climes but in the Tropics are used both indoors and outdoors. We have previously reported that diabetic patients may be unaware of the loss of slippers in the presence of severe DPN- the so-called Slipping Slipper Sign (SSS) [6]. This sign is said to be present when the response to the following question is in the affirmative: Have you ever lost a slipper without being aware of it? Since our first description in Trinidad [6], workers in Jamaica have verified our findings in an independent study [9]. This is a simple question which can identify those with feet most at risk.

Traditionally, NCS have been used to validate other screening tools in DPN and is regarded as the gold standard to evaluate other screening tests [10, 11]. In addition, ultrasonography is emerging as an alternative non-invasive technique for evaluating peripheral nerves [12 –14].

In this study we explored the relationships among patients with DPN according to presence of SSS with traditional clinical scoring i.e. Toronto Neuropathy Score and autonomic neu ropathy score, nerve conduction studies as well as ultrasonography of the right sural nerve.

METHODS AND SUBJECTS

Subjects

This was a prospective, cross sectional study among subjects with diabetes for which approvals were obtained from both the Ethics Committee of The University of the West Indies (U.W.I.), St Augustine, Trinidad and The South West Regional Health Authority, Trinidad. W.I.

Subjects were recruited from the diabetic clinic and wards of the San Fernando General Hospital, Trinidad. W.I. A diabetic subject was defined as one already diagnosed with and receiving anti-diabetic pharmacologic treatment. Other criteria for inclusion were that every subject had to be a member of either the outpatient clinic or one of the wards of the San Fernando General Hospital, wore slippers at least once a week, was greater than 18 years of age and was able to give written, informed consent. A slipper was defined as any footwear that did not carry a strap around the ankles, in keeping with the previously described definition [6].

Exclusion criteria were those with high alcohol consumption (>14 units/week for women and >21 units/week for men) [6]; use of illicit drugs, Vitamin B12 deficiency, hypothyroidism or any other known cause for peripheral neuropathy. Patients who were on hemodialysis, pregnant or had experienced foot ulcers, amputations or those who did not wear slippers were excluded.

Data collection

Demographic data including age, gender and self-reported ethnicity were recorded as well as duration of diabetes mellitus since diagnosis, medication use (oral hypoglycemic agents, insulin or both), history of cerebrovascular accidents, retinopathy, myocardial infarction, coronary bypass grafting, kidney disease and hypertension. Measurements of blood pressure using an automatic machine (Phillips Sure Sign V2+, supplier), weight while seated using an electronic scale (SOEHNLE Professional Wt. supplier) and waist circumference were also recorded [15–16]. Clinical questionnaires were administered on “The survey of Autonomic Symptoms” [17] and the first part of the Toronto Clinical Scoring System (TCSS) [18].

A lower limb examination [19] was performed in keeping with the TCSS. In summary, different modalities of sensation were tested at the sternum until the patient perceived the test stimulus reliably. Patients were then asked to close their eyes and light touch, pin prick, vibration, position and temperature were evaluated with a “yes” or “no” answer; (documenting 1 as correctly perceived and 0 if absent). Lower limb reflexes were also assessed and documented as present, reduced or absent and scored as 2, 1 or 0, respectively. Pressure sensation was evaluated using a 10 g monofilament on the dorsum of the big toe in a non-callused area. Temperature and pain sensation (using a neuro tip) were also assessed in a similar fashion. Vibration was tested using a 128-Hz tuning fork on the bony prominence of the dorsum of the big toe. The severity of DPN in these patients were ranked according to the TCSS [18] where less than or equal to 5 indicated no neuropathy, 6–8 indicated mild neuropathy, 9–11 indicated moderate neuropathy, and more than or equal to 12 indicated severe neuropathy.

Patients were further classified as having a positive SSS or negative SSS to their answers to the question: “Have you ever lost a slipper without being aware of it?” according to the description by Teelucksingh et al. [6].

A single neurologist (YSV), blinded to all clinical evaluations performed ultrasonography (Siemens Acuson X300) and NCS (Neurowerk EMG equipment) on all patients. Ultrasound assessment of the right sural nerves was performed documenting the maximal thickness [14] at the ankle (0 cm point) and leg (5 cm point) with the subject in a prone position using a GEi 12L-RS Linear Array Transducer of 13 MHz and exerted minimal pressure on the skin to prevent compression of the right sural nerve when obtaining measurements.

For NCS, the patient was again placed in a prone position and the antidromic technique was used, with recording from the ankle and stimulating in the foreleg. The sensory nerve action potential (SNAP) of the sural nerve was recorded (Neurowerk, SIGMA Medizin-Technik GmbH).

STATISTICAL ANALYSES

Data was analyzed using the SPSS version 12.0 for Windows. Normally distributed data was displayed as mean (standard deviation (SD)) and non-parametric data as median (interquartile range (IQR)). Dichotomous data were displayed as number (N) and percentage (%). Parametric data were compared by t-test or Pearson’s correlation, non-parametric data by Mann Whitney test or Spearman’s correlations and dichotomous data by Chi Square test as appropriate. Models were also examined with the nerve ultrasound data binary-coded about the median: (a) sural nerve leg diameter less than 3.7 mm and (b) sural nerve ankle diameter less than or equal to 3.2 mm. The NCS data were coded as zero i.e. absent or non-zero when a SNAP was detectable and amplitudes were recorded in millivolts.

Backward stepwise logistic regressions were performed with presence of the SSS as outcome variable with the following independent variables which all had significant binary relationship with SSS: duration of diabetes, sural nerve conduction (latency and amplitude were equivalent so only amplitude was used in the final model), sural nerve diameter (at leg and ankle), TCSS, Autonomic Symptom Score, self- reported retinopathy and positive history of cerebrovascular accident.

RESULTS

Demographic and physiologic data

Of the 87 patients invited to participate, 74 showed up on the days of the study. The reasons and characteristics of those who defaulted were not recorded.

Table 1 shows that the mean (SD) age of the participants was 60.2 (10.3) years. There were 21 (28.4 %) males and 53 (71.6%) females in the study; 53 (71.6%) were East Indians, 10 (13.5%) Africans and eleven classified as other (14.9%). Fifty-eight (78.4%) had at least one co-morbidity; 52 (70.3%) were being treated for hypertension, 8 (10.8%) had reported a previous cerebrovascular accident and 19 (25.7 %) gave a history of retinopathy. The median (IQR) duration of diabetes was 10.5 (5.8–23.0) years.

Table 1

Showing Patient Characteristics, Nerve Conduction Studies and Sonographic measurements of Maximal Thickness of the right sural nerve in 74 patients with Diabetes Mellitus

Variables	Mean	SD
Age (years)	60.2	10.3
Male Sex	21	28.4
Female Sex	53	71.6
Systolic BP (mmHg)	151.5	23.6
Diastolic BP(mmHg)	77.6	10.6
	Median	IQR
Duration of Diabetes (years)	12.5	5.8–23.0
Action Potential Amplitude (μv)	0.0	0.0–7.2
Sural Nerve Ankle Maximal Thickness (mm)	3.2	2.8–3.7
Sural Nerve Leg Maximal Thickness (mm)	3.7	3.1–4.1
Autonomic Neuropathy Score	5.0	1.8–9.0
Peripheral Neuropathy Toronto Score	8.0	5.0–12.0
	N	%
DM Oral Medication used	44	59.5
Insulin	46	62.2
Insulin and Oral Medication	28	37.8
Ethnicity - East Indians	53	71.6
Ethnicity -Africans	10	13.5
Ethnicity -Other	11	14.9
At least one comorbidity	58	78.4
Hypertension	52	70.3
Cerebrovascular accident	8	10.8
Retinopathy	19	25.7
Myocardial Infarction	7	9.5
Coronary Artery Bypass Grafting	2	2.7
Chronic Kidney Disease	0	0

Abbreviations: BP, Blood Pressure, SD, Standard Deviation, IQR, Interquartile Range, mm, millimetres, cm, centimetres, DM, Diabetic Medication, N, numbers, %, percentage.

Table 2 illustrates the comparison of variables between those with and without the SSS. It can be seen that there were no statistically significant differences between the groups for age, gender, ethnicity, hypertension, medication usage (insulin /oral hypoglycemic agents), weight nor waist circumference.

Table 2

Showing values of key baseline and outcome variables in patients with and without the Slipping Slipper Sign

Variable	Slipping Slipper	Absent	Slipping Slipper Present		p
Mean	SD	Mean	SD
Age	59.3	10.9	61.1	9.7	0.455
Systolic BP(mmHg)	150.9	19.6	152.2	27.1	0.812
Diastolic BP(mmHg)	77.7	10.4	77.6	11.0	0.964
Waist Circumference(cm)	99.4	11.8	96.9	14.0	0.414
	Median	IQR	Median	IQR	p
Duration of Diabetes (years)	10.0	4.0–20.3	15.0	8.5–25.0	0.028
AP Amp (μv)	4.0	0.0–10.8	0.0	0.0–0.0	0.002
Sural Nerve Maximal Thickness at ankle (mm)	3.5	3.0–3.9	3.0	2.3–3.4	0.002
Sural Nerve Maximal Thickness at Leg (mm)	3.9	3.3–4.2	3.4	2.7–3.8	0.007
Autonomic Neuropathy Score	3.0	0.0–7.8	6.0	2.5–14	0.001
Peripheral Neuropathy Toronto Score	5.0	2.5–7.8	11.0	8.5–13	0.000
Weight (kg)	71	66.2–83.5	76.4	59.6–92.4	0.620
	N	%	N	%
DM Oral Meds used	20	55.6	24	63.2	0.506
Insulin used	22	61.1	24	63.2	0.856
Insulin and Oral Medication used	6	16.7	10	26.3	0.314
Male Sex	11	30.6	10	26.3	0.686
Female Sex	25	69.4	28	73.7	0.686
East Indian Ethnicity	25	69.4	28	73.7	0.686
Non East Indian Ethnicity	11	30.6	10	26.3	0.686
At least one comorbidity	27	75.0	31	81.6	0.492
Hypertension	26	72.2	26	68.4	0.721
Cerebrovascular accident	1	2.8	7	18.4	0.030
Retinopathy	5	13.9	14	36.8	0.024
Myocardial Infarction	3	8.3	4	10.5	0.074
Coronary Artery Bypass Grafting	1	2.8	1	2.6	0.969

P, significance, SD, standard deviation, cm, centimeters, IQR, interquartile range, mv, millivolts, mm, millimetres, N, numbers, %, percentage.

On the other hand, those with SSS were more likely to have had diabetes for a far longer duration (15 vs 10 years, p < 0.028), more likely to have reported a previous diagnosis of retinopathy (p < 0.024) and cerebrovascular accident, p < 0.03) and there was a borderline significance with having previously experienced a myocardial infarction, p = 0.07.

In addition, Table also 2 shows that the sural nerve thickness in those with SSS was markedly smaller. Thus, the maximal thickness of the right sural nerve was (median (IQR)), (3.0 (2.3–3.4) mm vs 3.5 (3.0–3.9) mm, p = 0.002 and 3.4 (2.7–3.8) mm vs 3.9 (3.3–4.2) mm, p = 0.007) at the ankle and leg respectively, in patients with compared with those without the SSS. Table 3 shows that a significant negative correlation of 0.31 and 0.36 for the presence of SSS and sural nerve thickness at both leg and ankle respectively. Similarly, the amplitude of the revealed marked diminution in patients with the SSS compared to those without the SSS were (median (IQR) = 0.0 (0.0–0.0) μv vs 4.0 (0.0–10.8) μv, p = 0.002, Spearman’s correlation = –0.37, p < 0.001). All of the of patients with SSS had absent SNAP vs 22 % in patients without SSS.

Table 3

Showing Spearman’s Correlation showing associations between individual variables

	AP (Amp)	SN (Ankle)	SN (Leg)	AN	PN
AP Amp (μv)	0.94^†
SN (ankle) MT (mm)	0.35^†
SN (leg) MT (mm)	0.41^†	0.93^†
AN Score	–0.26	–0.17	0.09
PN/TCSS	–0.45^†	–0.42^†	–0.37^†	0.3^†
Slipping Slipper Sign Present	–0.37^†	–0.36^†	–0.31^†	0.38^†	0.63^†
At least one Comorbidity	–0.11	–0.32^†	–0.34^†	–0.04	0.05

^*p value < 0.05. ^†p value < 0.01. AP, Action Potential, Lat, latency, Amp, Amplitude, SN, Sural Nerve, AN, Autonomic Neuropathy, PN/ TCSS Toronto Clinical Scoring System, Peripheral Neuropathy, EI, East Indian, MT, Maximal Thickness, mm, millimetres.

Patients with the SSS had higher scores for the Toronto Neuropathy Score compared with those without, (Median (IQR) = 11.0(8.5–13) vs 5.0(2.5–7.80), p = 0.000) and Autonomic Neuropathy Score (Median (IQR) = 6.0 (2.5–14) vs 3.0 (0. –7.8), p = 0.001).

Relationships with nerve measurements

SNAP were more likely to be absent in those patients with a longer history of diabetes and those on insulin (P < 0.05 in all cases). Age was inversely correlated with maximal thickness of the sural nerve, p < 0.05, for both ankle and leg. Sural nerve diameter was lower in those with older age, longer duration of diabetes and diabetic patients with at least one comorbidity (p < 0.05 in all cases).

Table 3 shows that all measures of nerve involvement (NCSE, diameter) and the clinical questionnaire scores were significantly related to the presence of SSS though the correlation coefficients were moderate. The table also shows that the NCS and diameter parameters were also significantly correlated with the peripheral nerve TCSS (p < 0.01) but not with the autonomic neuropathy score. The presence of at least another additional comorbidity was associated with smaller sural nerve diameters at both the ankle and leg.

Questionnaire scores and presence of SSS

The TNSS peripheral neuropathy score significantly correlated with the presence of SSS (Spearmann’s Correlation = 0.63, p < 0.01) and was greater in those with higher systolic blood pressure and a longer duration of diabetes (p < 0.05). The Autonomic Neuropathy score also correlated with peripheral neuropathy score and the presence of SSS (Spearmann’s Correlation = 0.38, p < 0.01).

Multivariate analysis

Table 4 shows the results of backward stepwise logistic regression of variables with slipping slipper sign as outcome variable and with the following independent variables: sural nerve maximal thickness at the leg, the latency of the sural nerve, the Toronto Neuropathy Clinical Score, the Autonomic Symptom Score, retinopathy and cerebrovascular accident as independent variables. Of these only three factors had significant association with the presence slipping slipper; the Toronto Clinical score, Autonomic neuropathy Score and retinopathy.

Table 4

Showing Multivariate Analysis with Slipping Slipper Sign Present as Outcome Variable

Variable	b	p
Variable in the Final Model
Constant	–4.263	<0.001
Autonomic Neuropathy Score	0.116	0.040
Peripheral Neuropathy Toronto Score	0.385	0.000
	OR	p
Retinopathy Absent	6.80	0.017
Variables not in Final Model
	Score	p
Duration of Diabetes (years)	0.225	0.635
Diameter of Sural Nerve (Leg)<3.7 mm	0.018	0.894
Diameter of Sural Nerve (Ankle) < = 3.2	0.342	0.559
NCS: Latency of Sural Nerve = 0	0.463	0.496
Cerebrovascular Accident	2.190	0.139

B, regression coefficient, p, significance, OR, odds ratio.

DISCUSSION

Neuropathy is a frequent complication of diabetes mellitus and among other factors is related to duration and degree of control of glycemia [1]. It has a long course and because of its insidious nature symptoms may be overlooked by the patient or minimized by the busy clinician. For these reasons, it is not surprising that the prevalence of neuropathic symptoms in the primary care setting in Trinidad was relatively high at 49% [2] and consistent with prevalence figures (23–54 %) reported elsewhere in the world [3]. Those with severe neuropathy are at a 7- fold increased for foot ulceration [20], infection and up to a 15-fold risk of lower extremity amputation [21].

In the Tropics and sub-Tropics, footwear that carry no straps at the ankles [8] is frequently used. Because of this design element they can be slipped on and off easily. In non-neuropathic feet, this loss will be promptly recognised. With insensate feet, inadvertent slippage may occur and go unrecognised for prolonged periods. In such cases, motor neuropathy leads to loss of the slipper from the foot and sensory neuropathy leads to a delayed recognition of its loss- often to the amusement of both patient and observers. Amusing accounts of such events have been given by both patients and relatives in which individuals have inadvertently left slippers behind in taxis and buses only to realize this having arrived home at the end of the journey!

Our previous research has identified that the SSS identifies a sub-group of individuals with severe diabetic peripheral neuropathy and at high risk for foot ulceration and infection [6]. Not surprisingly, those with SSS also display high prevalence of other microvascular complications viz., retinopathy and nephropathy [22]. The current study objectively confirms that the sural nerve in SSS displays very abnormal nerve conduction characteristics (absent SNAP). In addition, sural nerve diameters are markedly reduced in SSS. These features are consistent with very severe neuropathy. A study done by Riazi et al. [14], showed findings similar to those demonstrated in our study in which that amplitude was smaller and latency prolonged in patients with diabetes.

Not surprisingly, patients with the SSS had a longer duration of diabetes and also had greater severity of neuropathy as judged by the Toronto Clinical Neuropathy Score (TCNS). It therefore appears that the SSS is an integral of many factors including prolonged duration and poor control of diabetes.

Multivariate Analysis showed a significant independent correlation with the presence of the SSS with autonomic neuropathy, Toronto neuropathy score and retinopathy. We have previously reported the latter [6]. In the current study, it is noteworthy note that there is also an association between Diabetic Autonomic Neuropathy and the SSS. It is well-known that diabetic autonomic neuropathy is a serious complication of diabetes and carries increased mortality risk [23]. Interestingly, patients with SSS were also more likely to have reported a prior cerebrovascular accident- a macrovascular complication associated with diabetes. In summary, this study highlights the strong association between the SSS and very abnormal peripheral nerves.

Patients with SSS also show evidence of advanced macrovascular disease. As such any patients in clinical practice who answers in the affirmative to the question: “Have you ever lost your slipper and not been aware of it?”, should be identified as having severe DPN and likely other associated complications of diabetes mellitus. The presence of SSS is easy to identify and thus can be utilized by non-medical staff. It can be used as a strategy to screen and bring to the fore those patients who may be completely unaware of their underlying neuropathy.

The SSS is not a diagnostic test for DPN but a predictor of those with objective vidence of a very severe form of this complication of diabetes. It remains to be seen whether screening for the SSS to detect insensate feet will impact future diabetic foot complications.

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

References

Jaiswal

, Divers

, Dabelea

, Isom

, Bell

, Martin

, et al. Prevalence of and Risk Factors for Diabetic Peripheral Neuropathy in Youth With Type 1 and Type 2 Diabetes: SEARCH for Diabetes in Youth Study. Diabetes Care. 2017;40(9):1226–32.

Gulliford

and Mahabir

. Diabetic foot disease and foot care in a Carribbean community. Diabetes Res Clin Pract. 2002;56(1):35–40.

Garcia

, Velaseo

, Cabrerizo

, Pe’rez

and Gomez

. Prevalence of distal diabetic polyneuropathy using quantitative sensory methods in a population with diabetes of more than 10 years’ disease duration. 2010;57(9):414–20.

Bansal

, Gudala

, Muthyala

, Esam

, Nayahallu

and Bhansali

. Prevalence and risk factors of development of peripheral diabetic neuropathy in Type 2 diabetes mellitus in a tertiary care setting. Journal of Diabetes Investigation. 2014;5:714–21.

Callaghan

, Kerber

, Langa

, Banerjee

, Rodgers

, McCammon

, et al. Longitudinal patient-oriented outcomes in neuropathy: Importance of early detection and falls. Neurology. 2015;85(1):71–9.

Teelucksingh

, Ramdass

, Charran

, Mungalsingh

, Seemungal

and Naraynsingh

. The slipping slipper sign: A marker of severe peripheral diabetic neuropathy and foot sepsis. Postgrad Med J. 2009;85:288–91.

International Diabetes Federation North America and Caribbean Region. IDF. [Homepage on the Internet]. Brussels, Belgium: International Diabetes Federation; 2014 [Cited 2019 June 19]. Available from: www.idf.org/.

Slipper. Merriam-Webster [Homepage on the Internet]. Massachusetts: Merriam Webster Dictionary; [Cited 2019 June 19]. Available from: www.Merriam-Webster.com/.

Gayle

KAT

, Tulloch-Reid

, Wilks

and Ferguson

. The slipping slipper sign: A simple test with high specificity and positive predictive value for peripheral neuropathy among diabetic patients. Clinics and practice. 2012;18:48–56.

10.

Cornblath

, Chaudhry

, Carter

, Lee

, Seysedadr

, Miernicki

, et al. Total neuropathy score: Validation and reliability study. Neurology. 1999;53:1660–4.

11.

Feldman

and Stevens

. Clinical testing in diabetic peripheral neuropathy. Can J Neurol Sci. 1994;21:S3–7.

12.

Watanabe

, Sekine

, Nishimura

, Kato

, Takeda

, Seishima and Matsuoka

. Sonographic Evaluation of the Peripheral Nerve in Diabetic Patients. J Ultrasound Med. 2010;29:697–708.

13.

Liu

, Zhu

, Wei

, Bao

and Hu

. Preliminary Evaluation of the Sural Nerve Using 22-MHz Ultrasound: A New Approach for Evaluation of Diabetic Cutaneous. PLoS ONE. 2012;7(4):e32730.

14.

Riazi

, Bril

, Perkins

, Abbas

, Chan

VWS

, Ngo

, Lovblom

, El-Beheiry

, et al. Can Ultrasound of the Tibial Nerve Detect Diabetic Peripheral Neuropathy? A Cross-Sectional Study. Diabetes Care. 2012;35:2575–9.

15.

Nishida

, Ko

and Kumanyika

. Body fat distribution and noncommunicable diseases in populations: Overview of the 2008 WHO Expert Consultation on Waist Circumference and Waist–Hip Ratio. European Journal of Clinical Nutrition. 2009;64(1):2–5.

16.

, Yang

, Shi

, Hsieh

, Hung

, Chiu

, et al.

Measurement of Waist Circumference Midabdominal or iliac crest?

Diabetes Care. 2013;36(6):1660–6.

17.

Zilliox

, Peltier

, Wren

, Anderson

, Smith

, Singleton

, et al. Assessing autonomic dysfunction in early diabetic neuropathy: The Survey of Autonomic Symptoms. Neurology. 2011;76(12):1099–105.

18.

Bril

and Perkins

. Validation of the Toronto Clinical Scoring System for Diabetic Polyneuropathy. Diabetes Care. 2002;25(11):2048–52.

19.

Perkins

, Olaleye

, Zinman

and Bril

. Simple Screening Tests for Peripheral Neuropathy in the Diabetes Clinic. Diabetes Care. 2001;24(2):250–6.

20.

Young

, Breddy

, Veves

and Boulton

AJM

. The Prediction of Diabetic Neuropathic Foot Ulceration Using Vibration Perception Thresholds: A prospective study. Diabetes Care. 1994;17(6):557–60.

21.

Bild

, Selby

, Sinnock

, Browner

, Braveman

and Showstack

. Lower-Extremity Amputation in People With Diabetes: Epidemiology and Prevention. Diabetes Care. 1989;12(1):24–31.

22.

Teelucksingh

, Ramdass

, Ramnath

, Teelucksingh