diabetes mellitus - diabetic neuropathies - ankle-brachial index
diabetes mellitus - neropatias diabéticas - índice tornozelo-braquial
According to the International Diabetes Federation (2015), there are more than 415
million people with diabetes worldwide and the prevalence of obesity and sedentary
lifestyle will increase this number. The projection of the International Diabetes
Federation is that in 2040 this estimate will be 642 million people affected by this
disease, and the risk of cardiovascular events and neurological complications is four
times that of a non-diabetic[1].
Population-based studies reveal that 50% of patients over 60 years of age develop
neuropathy in the early stages of type 2 diabetes, while neurological symptoms in
type 1 diabetes are usually manifested later[2].
Diagnoses of diabetic neuropathies rely on the characterization of symptoms and clinical
signs, such as pain being the reason for 40% of patients visiting a health service
annually, and 20% of these had felt pain for more than six months. Due to its complexity,
managing chronic pain is a challenge to the clinician[3].
The diagnosis of diabetic neuropathy is based on the characterization of the most
typical symptoms and clinical signs, such as pain and neurological conduction tests[4].
Peripheral arterial disease often affects younger patients and develops in 10% of
newly-diagnosed cases of diabetes. For this reason, it is necessary to be able to
evaluate arterial disease in a simple and quick way.
The ankle-brachial index (ABI) shows the existence of peripheral arterial disease
and is also related to cardiovascular mortality in diabetic patients, even in those
with coronary artery disease at an early stage[5].
Chronic complications render many patients unable to work[2]. Population-based studies of diabetics reveal that 50% of patients over 60 years
of age develop neuropathy, with peripheral diabetic neuropathy[6].
Diabetes mellitus is the leading cause of peripheral neuropathy, commonly manifested
as symmetrical and distal polyneuropathy.
A chronic increase of glucose leads to glycosylation of proteins and the final products,
which accumulate in the tissues, produce microvascular disease by direct deposition
of these proteins in endothelial cells or by generation of oxidative stress[7].
Vascular pathogens have been postulated and hypoxia and ischemia are also involved
in diabetic polyneuropathy. On a macroscopic level, the study of the distribution
and fiber loss in diabetic nerves also suggests a vascular disorder[8].
The diagnosis of clinical pain in peripheral diabetic neuropathy is based on the description
provided by the patient. The symptoms are distal and symmetric, with frequent nocturnal
exacerbation. Symptoms are typically described as pins and needles, deep pain, electric
shock, tingling, and burning sensations, and may present as hyperalgesia, allodynia,
or both[9].
Painful diabetic neuropathy may affect up to 50% of patients who have been diagnosed
for more than 25 years. The associated risk factors include age, genetic predisposition,
inflammation, oxidative stress, lipotoxicity, and glucotoxicity[10],[11]
Distal sensory polyneuropathy is the most common form in diabetes, and premature symptoms
are decreased sensation of vibration, touch, and the ability to feel a needle or monofilament.
There is evidence that both symptomatic and asymptomatic lesions may be present in
diabetic arterial disease, and these predict cardiovascular lesions[12].
The ABI is used to diagnose peripheral arterial disease, as well as to estimate future
progression and severity[13].
In the non-diabetic population, the incidence of ABI changes varied about 8.7%, and
the prevalence of 62.2% artery disease was very likely among patients with chronic
kidney failure, tobacco use and cardiovascular disease[14].
Normal levels are considered to be an ABI of 0.9 – 1.3; below this there is an arterial
insufficiency and high cardiovascular risk. It is recommended that the ABI is performed
on all patients older than 50 years of age or any patient with symptoms of pain when
walking[2],[13],[15].
Many individuals with diabetes have few symptoms because, despite extensive loss of
vascularized tissue, sensation loss is common due to the distal sensory polyneuropathy.
The artery disease mortality and morbidity is a common cause of hospitalization, where
the risk of ulceration is 15-25%[16].
The arterial disease, a risk factor irrespective of ulceration and amputation, is
present in 50% patients with ulcers, so early detection is of great concern[16].
The International Working Group on the Diabetic Foot Guidelines suggests that, added
to the symptoms of neuropathic pain, vascular insufficiency can be evaluated by a
manual Doppler and calculated by an ABI of less than 0.9, to assess for possible revascularization.
This peripheral arterial involvement is diffuse and particularly severe in the tibial
arteries with a high prevalence of occlusion[13],[17],[18],[19].
We believe that this study is of real importance, since through the application of
a simple examination, the ABI, along with an examination for neuropathic pain and
the DN4 questionnaire, further evidence of the relationship between arterial damage
and the presence of neuropathy in diabetics can be found.
METHODS
This cross-sectional study includes a descriptive analysis of a group of patients
with diabetes. The group consisted of patients with type 1 diabetes and type 2 diabetes,
who had previously been examined and diagnosed by a physician with experience in treating
diabetes (the author).
The DN4 questionnaire included 10 objective questions regarding sensitivity symptoms
related to painful peripheral neuropathy, requiring direct “yes” or “no” answers.
The experts examined patients with chronic, central neurological, or peripheral pain
and differentiated them by whether or not they presented with somatic tissue injuries.
This differentiation was based on the DN4, which includes pain descriptors and neurological
examination[19].
The diagnostic sensitivity of this questionnaire for neuropathic pain is 82.9% and
the specificity is 89.9% when patients respond in the affirmative to at least four
items[20].
The DN4 interview is discriminatory, suggesting its feasibility for large-scale epidemiological
studies, consistent with a major review on neuropathic pain[4].
Routine use of the ABI has been proposed for use in diabetic patients at high risk
of arterial disease. The Doppler method is a well-established and non-invasive technique
to assess peripheral arterial disease[15].
The ABI should be measured in the supine position after five minutes of rest. The
ABI reference scores are: a) > 1.3 = hardened vessels; b) from 0.9–1.3 = normal arterial
flow; c) < 0.9 = arterial occlusion. The ABI measurement is recommended for all patients
with diabetes over 50 years of age, or those presenting with symptoms of peripheral
arterial disease or other cardiovascular risk factors[8],[16],[17].
Patients’ DN4 scores were then subjected to ABI measurement, which was calculated
by dividing the highest systolic blood pressure in the lower and upper limbs[17].
Sampling
The study population included 225 patients with diabetes, recruited through the Diabetics
Association, who were then separated into groups of type 1 diabetes (n = 75) and type
2 diabetes (n = 150). Forty-six men and 29 women had type 1 diabetes, and 47 men and
103 women had type 2 diabetes.
For patients younger than 60 years, there were 60 patients in the type 1 group, and
37 in the type 2 group, totaling 97 patients. For those over 60 years of age, 15 had
type 1 and 113 had type 2 diabetes, totaling 128 participants, randomly selected from
the endocrinology outpatient clinic of the Diabetes Association.
According to the responses obtained on the DN4, when four responses were positive,
the patients were assessed by their ABI scores. Patients with less than four positive
responses were excluded; as well as smokers and those with renal disease.
The prevalence of neuropathy was considered when calculating the sample size of 225
patients, yielding a sample power of 0.9 with a sample error of 0.05.
This study was approved by the Hospital Research Ethics Committee under protocol number
451.639.
Statistical analyses
Statistics were analyzed using the Stata 9.2 software. The variables for analyses
were obtained from the DN4 questionnaire applied in this study. The percentage of
results obtained in the DN4 and ABI were determined in the two groups of diabetics.
The two groups were compared through RxC contingency or ANOVA tables, as indicated.
Chi-square, Fisher’s exact test, student’s t tests, and the Mann-Whitney tests were
used to assess the statistical significance of data. Differences between the two groups
were described in terms of odds ratios, and adjustments of ABI and DN4 interpretation
correlations based on the variables, using logistic regression models specifically
designed to meet the aims of the analyses. Differences with p-values less than 0.05
were considered statistically significant.
RESULTS
As shown in [Table 1], an ABI < 0.9 was associated with 43 type 1 patients and 89 type 2 diabetic patients.
Table 1
ABI and diabetes.
|
ABI
|
Type 1 diabetes
|
Type 2 diabetes
|
Total
|
|
< 0.9
|
43
|
89
|
132
|
|
0.9–1.3
|
30
|
51
|
81
|
|
> 1.3
|
2
|
10
|
12
|
Normal range for ABI = 0.9 - 1.3; Risk of obstructive arterial disease < 0.9 and >1.3.
ABI: ankle-brachial index.
On the other hand, an ABI > 1.3 in two type 1 diabetic patients, and one type 2 diabetic
patients was associated with a higher cardiovascular risk.
A normal ABI was found in 30 patients with type 1 diabetes and 51 patients with type
2 diabetes.
[Table 2] shows that altered ABI scores were observed in 58 individuals below 59 years of
age and in 86 individuals above 60 years of age. Normal ABI scores were observed in
39 patients with diabetes who were below 59 years of age and in 42 patients with diabetes
who were above 60 years of age.
Table 2
ABI score and age.
|
Age
|
Altered ABI
|
Normal ABI
|
Total
|
|
< 59 years
|
58
|
39
|
97
|
|
> 60 years
|
86
|
42
|
128
|
Altered and normal ABI values according to the age of participants with diabetes.
ABI: ankle-brachial index.
Scores corresponding to positive responses on the DN4 were combined with the altered
or normal ABI scores accordingly.
[Table 3] shows the ABI results allowing the classification of two possible situations, which
included individuals diagnosed with type 1 diabetes for either more, or less, than
10 years. A non-significant statistical correlation was obtained for this group, with
a p-value of 0.263.
Table 3
ABI and Type 1 diabetes disease duration.
|
Duration
|
Altered ABI
|
Normal ABI
|
p
|
|
< 10 years
|
12
|
5
|
|
|
> 10 years
|
27
|
22
|
0.263
|
ABI values and duration of type 1 diabetes, statistical significance. ABI: ankle-brachial
index.
[Table 4] shows patients with type 2 diabetes who presented with altered ABI scores, which
is in agreement with the observed variations in the duration of time since the diagnosis.
Table 4
ABI and Type 2 diabetes disease duration.
|
Duration
|
Altered ABI
|
Normal ABI
|
p
|
|
< 10 years
|
23
|
21
|
|
|
> 10 years
|
69
|
25
|
0.014
|
ABI values and duration of type 2 diabetes, statistical significance. Significant
p value. ABI: ankle-brachial index.
On the other hand, normal ABI scores were obtained by 21 patients who had been diagnosed
for less than 10 years and for 25 patients who had been diagnosed for more than 10
years. Comparison of these groups yielded a statistically significant correlation,
with a p-value of 0.014.
DISCUSSION
Previous literature indicated that neurological symptoms are reported by 38% of women
compared to 31% of men. However, fewer women presented with peripheral diabetic neuropathy
confirmed by neurological examination, and the risk of developing symptoms associated
with neurological pain remained 50% higher in women than in men[20].
A similar gender relationship was not observed in type 1 diabetes, contrary to the
positive responses observed from men with type 2 diabetes (46 men; 29 women). However,
a higher incidence of neuropathic pain was observed among individuals > 60 years of
age (113 in type 2; 15 in type 1 diabetes) according to the DN4 questionnaire. This
is consistent with other literature that reported a higher number of complications
with increases in the duration from diagnosis, and age, of patients[19].
In participants aged > 60 years, an ABI ≥ 1.3 was observed as a result of arterial
hardening. These scores were obtained in 10 patients with type 2 diabetes and in two
in the same age group with type 1 diabetes. This suggested a correlation between ABI
scores indicative of health complications and advancing age.
The group was diagnosed with sensory neuropathy, and peripheral arterial disease was
detected in almost 40% of the patients. Arterial calcification also develops in other
conditions associated with peripheral diabetic neuropathy, although it does not increase
at the same rate as it does in patients with diabetes[20],[21].
Diabetic arteriopathy involves the most distal vessels of the lower limbs, such as
the tibioperoneal, posterior tibial, and dorsalis pedis arteries[17],[18].
Neuropathic pain (as assessed in the DN4 questionnaire) directly correlated with altered
ABI scores, and was observed in the majority of our population (i.e., in 144 individuals;
64.2%).
Previous studies indicate that some symptoms are preferentially manifested in neuropathic
pain. In particular, paresthesia and dysesthesia are highly specific, and are used
as diagnostic indicators[22],[23].
Diagnosis becomes difficult when peripheral arterial disease is associated with peripheral
diabetic neuropathy, since the latter could mask claudication and pain at rest induced
by severe ischemia, with the consequent prevalence of neuropathic pain diagnoses.
It is also known that only one third of patients with diabetes present with intermittent
claudication[24].
However, the sensitivity of this ABI threshold (0.9–1.3) seems to be lower in complicated
type 2 diabetes, particularly in the presence of peripheral diabetic neuropathy; 13.6%
of patients with type 2 diabetes present with peripheral arterial disease compared
to 4% in the general population[15].
Altered ABI scores were observed in 27 patients with type 1 diabetes for over 10 years,
whereas normal ABI scores were observed in 22 patients in this group.
Altered ABI scores were observed in 12 patients with type 1 diabetes for less than
10 years, whereas normal ABI scores were observed in five patients in this group.
A p-value of 0.263 did not indicate a statistically significant difference.
A strong correlation between the time of diagnosis and the prevalence of altered ABI
scores was observed in our sample, when comparing diabetes diagnoses of more than
10 years.
The incidence of neuropathy and peripheral arterial disease increased by 5% per year
(on an average) in type 1 diabetes patients who had been diagnosed between 20 and
25 years. The risks of these complications are much higher 20 years after diagnosis,
increasing two- to three-fold beyond this period[25].
The role of improved glycemic control in the prevention of macrovascular disease in
diabetes is still uncertain. Premature death of these patients, as well as cardiovascular,
cerebrovascular, and peripheral arterial disease, limit the design of these studies
in both patients with type 1 and type 2 diabetes. Over the past 20 years, large-scale
studies, such as the Diabetes Control and Complication Trial for type 1 diabetes and
the United Kingdom Prospective Diabetes Study for type 2 diabetes, have reported premature
deaths in these populations, caused by diabetic arterial disease[26].
When evaluating the disease duration in type 2 diabetes, 23 patients who had had the
disease for less that 10 years had an abnormal ABI and 21 patients had a normal ABI.
In patients with type 2 diabetes for longer than 10 years, 69 patients had low ABI
scores and the ABI score was normal in 25 patients.
Previous studies have also compared the prevalence of cardiovascular disease in patients
with or without types 1 and 2 diabetes. In the same age group, cardiovascular disease
occurred in 44% of patients with type 1 and 51% of patients with type 2 diabetes[26].
We observed that 23 patients with type 2 diabetes diagnosed for less than 10 years
presented with an altered ABI. Normal ABI scores were found in 21 individuals. In
patients with type 2 diabetes for longer than 10 years, 69 patients had altered ABI
scores, whereas 25 had a normal ABI. By comparing patients diagnosed with type 2 diabetes
for longer than 10 years, a p-value of 0.014 was obtained, indicating a positive correlation
([Table 4]).
Data from the literature confirm the high prevalence of cardiovascular complications,
such as peripheral arterial disease, evolving in type 2 diabetes patients with a duration
of longer than 10 years. Our data are consistent with these findings. We found that
increased age of patients with type 2 diabetes, as well as a diagnosis for more than
10 years, were considered risk factors[17],[18].
Therefore, the history of neuropathic pain and vascular insufficiency, which can be
evaluated by the Doppler method measuring the ABI, allows the International Working
Group on the Diabetic Foot Guidelines to be interpreted such that persons with peripheral
arterial disease symptoms and ABI scores < 0.9 should be assessed for possible revascularization[13],[14],[17].
Although ABI scores > 1.3 have no correlation with an accurate evaluation of ischemia,
they are associated with high cardiovascular risk, as calcification may occur with
normal ABI scores in the absence of clinical signs of peripheral arterial disease.
However, an ABI score < 0.9 indicates peripheral arterial disease, which should be
carefully confirmed by other complementary examinations[27].
Our data suggests that the populations with type 1 diabetes and type 2 diabetes might
increase if the DN4 questionnaire were administered after clinical assessment of sensory-motor
neuropathy. Our study participants had paresthesia and burning pain symptoms, although
claudication, tingling, and prickling sensations, as well as changes to cold perception,
were almost non-existent, thus hindering the minimum score of four positive responses
on the DN4. We hypothesize that a clinical evaluation by measuring ABI, followed by
the DN4 would alter our results.
This study aimed to show that neuropathic pain and peripheral diabetic neuropathy
occurs in patients with type 1 diabetes and type 2 diabetes, in combination with peripheral
arterial disease. This association was confirmed by the DN4 questionnaire, in which
altered ABI scores were observed in most of the patients who responded positively.
Very often, lower limb pain is attributed to peripheral neuropathy, while the possibility
of peripheral vascular disease is not investigated.
CONCLUSION
The present study emphasizes the appropriate clinical and semiological evaluation
of patients with diabetes, especially those who have been diagnosed for longer than
10 years. Moreover, it aimed to apply the DN4 questionnaire in all patients with type
1 diabetes and type 2 diabetes, even when the diagnoses occurred less than 10 years
prior, and to manually measure ABI scores in these patients once a year, using the
Doppler method (8 mHZ).
Our findings suggest that ABI measurement is necessary in all patients with type 1
diabetes and type 2 diabetes when peripheral diabetic neuropathy is suspected, but
also in asymptomatic patients diagnosed for longer than 10 years, regardless of neuropathic
pain or indicators of ischemia.
Neuropathic pain often occurs in the absence of any abnormalities in physical examinations,
such as hypoesthesia/anesthesia28.
The present study is clinically relevant because it investigates the ankle-brachial
index, in combination with a neurological examination and the presence of neuropathic
pain. This combination will contribute to the diagnosis of peripheral neuropathy and
the search for further evidence correlating arterial damage and neuropathy in patients
with diabetes.
