Evaluation of Different HbA 1 c Levels to Assess the Risk of Peripheral Neuropathy Among Type 2 Diabetic Patients Along With Other Conventional Risk Factors

This cross sectional study was carried out in the outpatient department of Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders, Bangladesh (BIRDEM) General Hospital, Dhaka, to explore the factors influencing or related to the development of the diabetic peripheral neuropathy (DPN) with specific concern to the HbA1c levels. A total of 400 patients with type 2 diabetic were selected to collect information on sociodemographic, blood pressure, anthropometry and lipid profile. Glycaemic status was assessed by HbA1c and plasma glucose levels. Prevalence of DPN was 16.8%, with male 10.8% and female 20.9%. Increased HbA1c categories above 7.0% were significantly associated with increased prevalence of DPN (9.2 Vs. 10.5 Vs 26.5%; χ = 19.446, p = .000). Logistic regression models showed that the risk of DPN was strongly increased at the HbA1c categories ≥8% (OR = 3.57; 95% CI: 1.75-7.26). Advanced age (OR = 1.97; 95% CI: 1.12-3.47), longer duration of diabetes (OR = 1.81; 95% CI: 1.02-3.19), lacking of physical exercise (OR = 2.60; 95% CI: 1.47-4.58), female gender (OR = 2.17; 95% CI: 1.21-3.89), fasting blood glucose (OR = 1.153; 95% CI: 1.058-1.255) and blood glucose 2 hours after breakfast (OR = 1.096; 95% CI: 1.029-1.168) were significant risk factors of DPN. However, there is need of a large scale community based prospective study to validate the results.


Introduction
Neuropathy is one of the most common vascular complications of diabetes that can cause significant morbidity and mortality. 1It is estimated from a comprehensive collection of epidemiologic studies that the prevalence of neuropathy in diabetes patients is approximately 30% in hospital patients and 20% in community patients. 2A commonly cited study in 1977 reported that roughly 7% of patients had neuropathy upon diagnosis of diabetes, and the incidence approached 50% for patients with diabetes for more than 25 years. 3However, it is difficult to accurately approximate the true prevalence of diabetic neuropathy, because the criteria for diagnosis vary, epidemiologic studies are limited to patients receiving medical care, and diabetes remains undiagnosed in a large population of diabetes patients. 4abetic neuropathies are heterogeneous with diverse clinical manifestations.They include focal, diffuse, sensory, motor and autonomic neuropathy.Most common among the neuropathies are chronic sensorimotor diabetic peripheral neuropathy (DPN).DPN accounts for approximately 75% of the diabetic neuropathies. 5t is defined as the presence of symptoms and/or signs of peripheral nerve dysfunction in people with diabetes mellitus (DM), after exclusion of other causes. 6It affects large and small afferent nerve fibers to varying degrees, resulting in mixed symptoms and sensoryloss. 1,7anifestations classically progress from the most distal extremities (the fingers and toes) in a symmetrical pattern that is generally described as a glove-and-stocking distribution.Typically, patients experience burning pain, electrical or stabbing sensations, paresthesia, hyperesthesia, and deep aching pain but sometimes they may experience simple numbness so it is important to realize that lack of symptoms does not rule out presence of neuropathy.Neuropathic pain is typically worse at night, and the symptoms are most commonly experienced in the feet and lower limbs, although in some cases the hands may also be affected. 7,8Upto 50% of DPN may be asymptomatic and patients are at risk of insensate injury to their feet. 5Up to 85% of amputations among diabetic patients are preceded by foot ulcers. 9Examination of the lower limb usually reveals sensory loss of vibration, pressure, pain, and temperature perception (mediated by small and large fibers) and absent ankle reflexes.Signs of peripheral autonomic (sympathetic) dysfunction are also frequently seen and include a warm or cold foot, sometimes with distended dorsal foot veins (in the absence of obstructive peripheral vascular disease), dry skin, and the presence of calluses under pressure-bearing areas. 8The prevalence of DPN varies in the literature from 5-100%, which may reflect the different diagnostic criteria and diverse study populations. 10e diagnosis of DPN can only be made after a careful clinical examination, and all patients with diabetes should be screened annually for DPN by examining pinprick, temperature, and vibration perception (using a 128-Hz tuning fork), 10-g monofilament pressure sensation at the distal halluces, and ankle reflexes.Combinations of more than one test have >87% sensitivity in detecting DPN.Loss of 10-g monofilament perception and reduced vibration perception predict foot ulcers.The feet should be examined for ulcers, calluses, and deformities, and footwear should be inspected.Nondiabetic causes of neuropathy including serum B12deficiency, hypothyroidism, uremia and chronic inflammatory demyelinating polyneuropathy should be excluded.A combination of typical symptomatology and distal sensory loss with absent reflexes, or the signs in the absence of symptoms, is highly suggestive of DPN.Different scoring systems have been developed for monitoring progression or response to intervention in clinical trials. 8erve conduction tests (NCS) may show reduced functioning of the peripheral nerves, but seldom correlate with severity of DPN and are not appropriate as routine tests for the condition 11 .The practitioner may wish to refer the more complex patient, or those in whom diagnosis needs confirmation, to a neurologist for specialized examination and testing. 8st investigators now agree that diabetic vascular complications result from the interaction of multiple metabolic, genetic and other factors.The primary risk factor for diabetic neuropathy is hyperglycemia. 2 The annual incidence of diabetic neuropathy in the Diabetes Control and Complications Trial (DCCT) was approximately 2% in conventionally treated patients, but that rate dropped to 0.56% in intensively treated type 1 diabetes mellitus patients. 12The UK Prospective Diabetes Study (UKPDS) 13 failed to support a similar correlation between the incidence of neuropathy and glycemic control in type 2 diabetes patients, but the progression of diabetic neuropathy is dependent on glycemic control in both type 1 and 2 diabetes patients, and the pathologies are considered similar. 2,3,5The DCCT and also UKPDS provide evidence that the risk for micro and macro vascular complications begins to increase at an HbA1c level of 6.5%. 12,133][14][15][16] Sabanayagam et al showed that increasing HbA1c categories had a higher prevalence of peripheral neuropathy. 17The duration of diabetes also increases the risk of neuropathy, but the association between duration and prevalence may depend in part upon patient age, which itself is a risk factor. 2,18Cigarette smoking, alcohol consumption, hypertension, retinopathy, obesity, microalbuminuria, height, hyperlipidaemia and hypercholesterolaemia are all considered as potential risk factors for diabetic neuropathy 2,18,19 .
Although it is known that the high blood sugar level and increase level of HbA1c results DPN there are still various open questions related to the HbA1c-based diagnosis.However, perhaps the most important open question is, how well does HbA1c predict complications.An identical statement was made in 2009 by the IEC on the role of the HbA1c. 20"The ultimate goal is to identify individuals at risk for diabetes complications so that they can be treated."Evidence-based recommendations for care of DPN is mainly based on trials in Anglo-Caucasian populations, but there is very little indication and randomized trial evidence of how well these guidelines are implemented in South Asia specially in Bangladesh.So far, there are a very few clinical studies regarding prevalence and risk factors of DPN 21 but no published data regarding the risk assessment of DPN by different levels of HbA1cin Bangladesh.Therefore, it was attempted to do a clinical study in this regard.The study was aimed to gain new insights into how various risk factors especially different cut-off points of HbA1c affect and reflect the risk of DPN among patients with type 2 diabetes which will provide necessary data to ensure improved preventive measures and care for diabetic patients.

Materials and Methods
This was a cross-sectional study conducted on 400 type 2 diabetic patients under follow-up attending the outpatient department (OPD) of BIRDEM General Hospital from January 2014 to December 2014.Type 2 diabetic patients diagnosed in accordance with the World Health Organization (WHO) criteria for both gender and age group were enrolled in the study.Subjects were of 30-60 years of age, with a duration of diabetes from 2 to 10 years.Patients with vitamin B12 deficiency, hypothyroidism, uremia, chronic hepatic diseases, chronic arthritis, alcoholism, cerebro-vascular disease, parkinsonism, uremia and acute or chronic musculoskeletal disorders, pregnant diabetic cases or gestational diabetes, type 1 diabetics and patients of haemoglobinopathies were excluded from the study.
BIRDEM is a 550-bed general tertiary level hospital with the most modern disciplines.The patients were registered in BIRDEM after confirmation of diabetes and pre-diabetes.During registration, each patient was given a 'reference number (Ref No)' with a 'diabetic guide book'.This registered medical book printed with specific 'Ref No' contains all baseline information of the patient and necessary advices related to diabetic management, advices and results of screening tests for different diabetic complications and blank pages for subsequent follow-up information.For subsequent follow-up which were given by doctors appointed in BIRDEM outdoor, patients come at a specified date and time written in diabetic guide book.Detection and management of different diabetic complications were done by specialized doctors in outdoor and they keep record in the guidebook.
In this study, a total of 400 type 2 diabetic patients were interviewed with pre-designed and pre-tested questionnaire, included information on sociodemographic (age, sex, family history of diabetes, geographical location, socioeconomic factor, educational history, occupational history), lifestyle characteristics (physical activity, smoking history etc.), blood pressure and anthropometry (height, weight, calculated body mass index) of the participants were collected.The duration of diabetes was also recorded.The selected patients were evaluated for the presence of peripheral neuropathy through reviewing the physicians' notes in the patients' medical report which were recorded in their diabetic guide book.They diagnosed the patients on the basis of symptom profiles and neurologic examinations.Symptoms profile includes burning, numbness, tingling, fatigue, cramping, aching feelings in the lower extremity (feetor in the calf), nocturnal exacerbation of the symptoms or present equally at day and night, the symptoms awake the patient from sleep and walking or standing maneuvers reduce the symptoms.For neurologic examination the procedure was explained and the tests applied on the patient's hand prior to the examination.The patient had to close the eyes during the examination of the Achilles tendon reflex by the broad end of the reflex hammer, vibration perception by biothesiometer, thermal sensation by cold sponge at the dorsum of the foot, and tactile sensation by pin-prick at the cuticle of the 1 st toe.In this study, 96 were DPN patients amongst the 400 type 2 diabetic patients.Of them, 11 patients were sent for nerve conduction study (NCS) to evaluate further complications of the disease.The glycaemic status of the participants were assessed by HbA1c, FBG and two hours ABF.In the current study participants were categorized into 3 groups by 3 HbA1clevel.These were good control group (HbA1c<7.0%),average control group (HbA1c 7-7.9%) and poor control group (HbA1c ≥8.0%).The participants were compared in these 3 HbA1c categories.HbA1c was measured by BIO-RAD variant which was modified HPLC method.Fasting lipid profile was also measured.Ethical clearance was obtained from the Institutional Ethical Committee of the BIRDEM Hospital.

Statistical analysis:
To compare the different variables among the groups, Chi-square tests were done for categorical data and Student's ttest for quantitative data.Univariate and multivariate logistic regression analysis were performed to estimate the casual effect of each predisposing factor on response variable i.e. development of peripheral neuropathy or remained normal.Odds ratio (OR) with 95% confidence interval (CI) were provided.All statistical tests were considered significant at a level of p < 0.05.SPSS software, version 21 was used for the statistical analysis.

HbA 1c categories (%)
HbA1c category ≥8% (OR = 3.57; 95% CI: 1.75-7.26)However, this difference may be due to diagnostic criteria used in those western studies differ from the current study, which was also suggested by Morkid and associates. 21The mean age of our subjects was 51.93 ± 6.33 years, which may confirm that the diabetes population in this part of the world is relatively young compared to the West. 21,28,29 was observed in this study that the advanced age and longer duration of diabetes were important and significant risk factors for peripheral neuropathy.These results were consistent with findings of other studies. 21,30,31It was to be found that the peripheral neuropathy was more common in females than males which was also observed by Alam 32 but contrast to other studies. 21,22,24,33This study also revealed a significant association of peripheral neuropathy with non-urban(rural and semi urban) residence.The higher prevalence of DPN in non-urban participants may be due to referral bias, as this center was offering advance services for diabetic complications and commonly complicated than non-complicated diabetic patients from rural and suburban area came here.It was observed that HbA1c, blood glucose levels in fasting and two hours after breakfast were important risk factors of peripheral neuropathy.These findings were consistent with other studies 30,34 .It was not observed any significant association of systolic and diastolic blood pressures with peripheral neuropathy.Ramachandra et al 35 also did not report any significant association of hypertension with diabetic neuropathy, but others were in contrast with this current study. 30,36Lack of physical exercise was also a contributing factor in this study.
Several studies indicated that HbA1c may show a glycaemic threshold with micro and macro vascular complications of diabetes, suggesting it may additionally be useful biomarker to identify individuals at risk for different vascular complications. 17,37,38In this study, it was observed that increasing HbA1c categories above 7.0% were significantly associated with increased prevalence of DPN.Logistic regression models of univariate analysis showed that the risk of DPN strongly increased at the HbA1c categories ≥8%.Even on multivariate analysis after adjusting potential confounding factors (advanced age, longer duration of diabetes, gender, hypertension, lacking of physical exercise) we found that HbA1c category ≥8% is important risk factors for peripheral neuropathy.
Results of this study were consistent with Sabanayagam et al 17 who reported that increasing HbA1c categories had a higher prevalence of any retinopathy, mild retinopathy, moderate retinopathy, CKD, micro or macro-albuminuria and peripheral neuropathy.Another study also found the similar association of HbA1c with retinopathy, nephropathy and neuropathy. 30oungas et al observed that for microvascular events the apparent threshold of HbA1c level was 6.5%. 38They also revealed that above thresholds, a higher level of HbA1c was significantly associated with higher risks of microvascular events in a log-linear manner.Below these thresholds, there was no significant relationship between mean HbA1c level and risks.In the current study there were relatively few DPN events observed at HbA1c levels less than 7.0% may be due to small sample size so it could not properly be evaluated the HbA1c levels below 7.0% in DPN.

Conclusion
Findings of this study suggest that increasing HbA1c categories above 7.0% were significantly associated with increased prevalence of DPN and the risk increases markedly at HbA1c levels ≥8%.The prevalence and risk of DPN also increased with advanced age, longer duration of diabetes, poor glycaemic control, lack of physical exercise, female gender etc. Careful assessment of the risk factors of DPN among diabetic patients, and control of HbA1c and appropriate preventive measures are thus recommended.

Table II
Clinical variables related to peripheral neuropathy

Table IV :
Relationship between peripheral neuropathy and HbA1c categories

Table V :
Multivariate analysis of risk factors for peripheral neuropathy.