Prevalence of peripheral neuropathy in type 2 diabetic adults in central Kerala: A cross-sectional study

Materials & Methods

A cross-sectional study was conducted in individuals aged 30 and above of both genders diagnosed with Type 2 Diabetes mellitus for at least one year and residing at the Ettumanoor municipality. The study was conducted from September 2020 to April 2021, over a period of eight months at the department of Community Medicine, Government Medical College (GMC), Kottayam, Kerala, India after obtaining approval from the Institutional Review Board of GMC, Kottayam, Kerala. Written informed consent was obtained for participation in the study and use of the patient data for research and educational purposes. The procedures in the study follow the guidelines laid down in the Declaration of Helsinki.

Consenting individuals aged 30 yr and above, of both genders, diagnosed with type 2 diabetes mellitus for at least one year were included in the study. Individuals with type 1 diabetes mellitus, gestational diabetes, past history of cerebrovascular accident, end-stage renal disease, hypothyroidism, peripheral occlusive vascular disease, and history of major spinal trauma or surgery were excluded from the study. Exclusion was done by verifying hospital records.

According to a study done at Chittoor, Andhra Pradesh, the prevalence of diabetic peripheral neuropathy (DPN) was estimated to be 39.3 per cent⁵. The sample size was estimated after taking the prevalence of above-mentioned study, and using the formula, Z² PQ/d². Relative precision taken here is 15 per cent of prevalence, thus the sample size obtained was 263. As cluster sampling was planned, after giving a design effect of 2, the minimum sample size was 263 x 2 =526.

Ettumanoor Municipality, with a total area of 24.78 square km according to the 2011 census, comes under Kottayam district, in the State of Kerala, and covers a population of 51,129. It consists of 35 wards, and the study participants were selected from all the wards. The number of persons chosen from each ward was 15 (526/35), and the houses were selected from each ward, by the following method.

A location near the centre of the ward was selected. A random direction, defined by spinning a bottle and choosing the direction that the bottle neck points to, was taken, and a random household, along the chosen direction pointing outwards from the centre of the community to its boundary, was chosen. The nearest household to the first one was chosen next and then houses were taken one after the other till the required sample in that cluster was met. Only one individual with diabetes was selected from each household.

A total of 543 individuals were approached for participation in the study. However, 17 declined to participate, and they were subsequently excluded from the study. A semi-structured interview schedule was administered to all the study participants, containing information on sociodemographic data, BMI, disease-related factors, treatment-related factors, and factors related to footcare. Modified Kuppuswamy classification was used to assess the socioeconomic status⁸. BMI is based on Asian classification, where underweight is <18.5, 18.5-22.9 is normal, 23-24.9 is overweight, and >/=25 is considered obese⁹. Smoking was measured in pack years. Current smokers were those who reported to smoke at least 100 cigarettes in their lifetime and who, at the time of the survey, used tobacco for smoking. Ex-smokers were those who reported to smoke at least 100 cigarettes in their lifetime and who, at the time of the survey, did not smoke at all. Never smokers were respondents who never smoked 100 cigarettes in their lifetime. A person who never consumed alcohol in their lifetime was considered anon-alcoholic, and occasional if they consumed alcohol 1-2 times/month, often if taken 1-2 times/wk, and very often >/=3 times/wk.

Treatment-related factors included irregular intake of medicines and self-modification of medicines. Participants who failed to take regular medications as prescribed more than half of the time over the last month were considered as on irregular medication. Those who altered their medications during the last month, like adjusting insulin doses, were considered to have done self-modification of medication. Fasting blood sugar (FBS) and postprandial blood sugar (PPBS) laboratory values recorded within the past two months were taken. Those with FBS values in the range of 80-130 mg/dl and PPBS <180 mg/dl were considered as controlled¹⁰.The presence of diabetes related complications like retinopathy and nephropathy was confirmed based on hospital records. DPN was screened using a previously validated tool, the Michigan neuropathy screening instrument (MNSI)¹¹. MNSI is a screening tool for DPN, developed and validated by the University of Michigan. Interview questions were translated into the local language and were administered by the principal investigator. Examination was also performed simultaneously at the participants’ houses by the principal investigator, who was trained in performing ankle reflex, assessing vibration sense using a tuning fork, and evaluating protective sensation using a monofilament.

MNSI has a sensitivity of 80 per cent and specificity of 95 per cent, and had two components, the history and the physical assessment component. The first part of the screening instrument consisted of 15 self-administered ‘yes or no’ questions on foot sensation, including pain, numbness, and temperature sensitivity. A score of 4 or above indicates neuropathic symptoms. The second part of the MNSI is a brief physical examination involving inspection of the feet for deformities and ulceration, assessment of vibration sensation at the great toe, grading of ankle reflexes, and monofilament testing. Patients screening positive on the clinical portion of the MNSI (greater than 2.5 points on a 10-point scale) are considered neuropathic¹².

The test of vibration was performed bilaterally using a 128 Hz tuning fork placed over the dorsum of the great toe on the distal interphalangeal joint. Zero score indicates intact vibration sensation; the examiner senses the vibration on his/her finger for <10 sec longer than the subject feels it over the great toe. Reduced sensation was scored as 0.5 if the examiner sensed vibration for 10 sec or more, and the lack of vibration sensation was scored as 1. Ankle reflex was assessed using a percussion hammer. If the reflex was absent, the Jendrassic manoeuvre was performed. Reflexes elicited without reinforcement were scored as zero, those with reinforcement scored as 0.5, and if absent scored as one. To assess protective sensation, a 10 g monofilament was placed perpendicular to the skin and pressure applied until the filament just buckled with a contact time of two seconds. The 10 g monofilament was applied over ten points on each foot; a ‘yes’ response indicated filament sensation. Eight correct responses out of 10 applications were considered as normal, 1-7 correct responses as reduced sensation, and no correct answer as absent sensation¹².

Results

This cross-sectional study was conducted among 526 individuals of Ettumanoor Municipality in Kottayam district. Study participants were selected from all 35 wards of Ettumanoor municipality with an equal number of subjects from each ward. The baseline socio-demographic variables (Table I).

The mean age of the study participants’ was 62.72 yr, with a standard deviation of 9.01 yr, and the age range was between 35 and 98 yr. The majority were females (66.3%) and belonged to the upper-lower socioeconomic status, according to the modified Kuppuswamy Scale¹³. In the study population, 402 individuals (76.4%) were diagnosed with hypertension, 276 (52.4%) had dyslipidaemia, 83 (15.7%) had coronary artery disease, 34 (6.46%) had osteoarthritis, 28 (5.32%) had asthma, 7 (1.3%) had COPD, and 11 (2%) had carcinomas

In the current study median duration of diabetes among the study participants is seven yr with an interquartile range of 10 yr. The prevalence of DPN among diabetic individuals aged 30 yr and above in Ettumanoor municipality, as per MNSI Examination score, was found to be 34.2 per cent (95% C.I.: 30.2-38.4) (Fig. 1).

Close

Fig. 1.

Distribution of the study participants (N=526) based on prevalence of diabetic peripheral neuropathy (DPN) using MNSI Examination score.

Export to PPT

Age, education, occupation, duration of diabetes, height more than 160 cm, use of insulin, irregular medication, and presence of hypertension, retinopathy, and nephropathy were found to have a significant association with DPN on univariate analysis (Table II). Gender, socioeconomic status, BMI, habits like smoking and alcohol consumption, FBS, PPBS, and barefoot walking did not have any significant association with DPN.

Those variables that were found to have a significant association with DPN with P<0.05 in univariate analysis were further analysed using binary logistic regression (Table III). The method of logistic regression used was the Enter method. Model summary as explained by the Negelkerke R² value obtained in the final model was 0.31. According to this model, the factors which were found to be independent predictors (with P<0.05) of DPN were individuals aged >60 yr [adjusted OR-1.88 (1.10-3.21)], educational status, high school and below [adjusted OR- 1.26 (1.12-1.43)], unemployment adjusted [OR-1.79 (1.08-2.97)], height >160 cm [adjusted OR-1.86 (1.15-3)] and insulin use [adjusted OR-1.45 (1.08- 1.94)].

Discussion

Among the 526 study subjects, an increased prevalence of DPN was observed in the age group of > 60 yr (66.7%) who were either retired or had stopped working. In most houses, the working population was not available during data collection time. Furthermore, because diabetes prevalence is thought to rise with age, most of the participants as expected in this study were elderly.

Socioeconomic status classification was done with Modified Kuppuswamy classification (Reference to CPI-March 2021), and a majority of the current study population belonged to the upper -lower class, possibly due to the change in Consumer Price Index (CPI) in March 2021 when compared to previous years.

In the current study, the prevalence of DPN among individuals aged ≥30 yr was found to be 34.2 per cent (95% CI- 30.2-38.4). Similar results were found in a study conducted by Darivemula et al⁵ in rural South India, where the prevalence of DPN was found to be 39.3 per cent. A community-based study to estimate the prevalence and associated factors of diabetes [Chennai Urban Rural Epidemiology Study (CURES-55)] found that DPN prevalence was 26.1 per cent¹⁴. Another study done at Uttar Pradesh by Gill et al¹⁵ showed a prevalence of 29.2 per cent.

Higher prevalence was obtained in other studies done in Kerala by George et al¹⁶, in which the prevalence of DPN was 47 per cent^, and by Sankar et al¹⁷, where the prevalence was 45.8 per cent. A study done in South India by Jasmine et al¹⁸, the prevalence was 44.9 per cent, and another study conducted at North India by Bhasker et al¹⁹, the estimated prevalence of DPN was 42 per cent. A much lower prevalence was obtained in a study conducted by Ashok et al²⁰ in South India, where it was 19.1 per cent. The prevalence of DPN across 14 countries in the INTERPRET-DD Study was 26.71 per cent⁶. These differences in the prevalence may be due to the difference in study settings and socio-demographic features, and the difference in the screening tool used. Most of the studies were hospital or clinic-based, only a few studies were community-based.

In this study, it was found that as age increased, there was a significant increase in the prevalence of DPN (P value=0.015). Similar results were obtained in other studies^5,7,14,21, where there was an increasing trend of DPN in older individuals. This may be due to biological changes leading to increased production of advanced glycosylation end-products (AGEs), defect in the polyol pathway, nerve vascular alterations, and impaired resistance to oxidative stress.

In the current study, there was no significant association between gender and DPN. Similar results were obtained in a community study (CURES-55)¹⁴ and a study done in Kerala²², where there were no gender specific differences. People having a high school education or below had a higher proportion of DPN (38.7%) (P< 0.001). Similar results were obtained in studies done by Kamalarathnam et al⁷ and Jasmine et al¹⁸. DPN was observed to be higher in unemployed (39.8%) when compared to employed (25.1%) a with a P< 0.001 in this present study similar to Kamalarathnam et al⁷ and Mathiyalagen et al²². For those participants with a duration of diabetes of more than 10 yr, a significant association was present (P=0.001). In a study done by Darivemula et al⁵ there was a significant association between the duration of diabetes of > 5 yr and DPN [(52.6% OR = 0.23 (0.13-0.39)P=0.01]. Similar results were obtained in the multicentric study, INTERPRET-DD⁶.

There was a significant association between increased height (above 160 cm) and DPN (P=0.01). Similar results were shown by other studies by Robinson et al², where peripheral neuropathy risk was significantly higher among those taller than 175.5 cm [(odds ratio=2.3,(95% CI: 1.5-3.5)] and Sankar et al²¹. In another study from Manipal²⁴, peripheral neuropathy increased at a height of more than 167 cm in males and at a height of > 159 cm in females.

About 53.2 per cent of participants on insulin therapy had DPN with a P<0.001 similar to earlier observations^21,25,26 (OR: 4.50; CI: 1.50-13.47). Hypertension was associated with DPN in the current study (P=0.004). Similar to other studies^5,26. The evidence from the UK Prospective Diabetes Study (UKPDS) also demonstrated that improvement in blood pressure reduced diabetes related complications²⁷. In the current study, smoking and alcohol had no significant association with DPN, while the presence of other microvascular complications (retinopathy and nephropathy) had a significant association similar to previous reports^4,5.

The strength of this study was that as it is a community-based study and the sampling used was probability sampling, generalisability was possible. But, as this was a cross-sectional study, the temporal association of factors with DPN could not be established. Also, there can be other causes of peripheral neuropathy among diabetics-like vitamin B12 deficiency, vitamin B1 deficiency, autoimmune neuropathies, and uremic neuropathy, which couldn’t be differentiated. Some of the variables in the study, like smoking, consumption of alcohol, treatment related factors, and footcare practices, were self-reported, with a chance of recall bias. The representation of the working group was lower in the current study. Also, as the study focused on subjective assessment of neuropathy, there is a chance for observer bias, MNSI examination part was done initially to overcome this.

Overall, the prevalence of DPN was found to be high in the study population. Factors which were found to be independent predictors of DPN include age > 60 yr, below high school education, unemployed status, height more than 160 cm, Insulin use and presence of nephropathy. The association of diabetes duration, irregular medication, self-modification of medications, presence of hypertension, and retinopathy with DPN was no longer significant after adjusting for confounders. All diabetics in the community should undergo screening for microvascular complications, including DPN, starting at diagnosis and annually thereafter. Early community-based diagnosis using screening tools such as MNSI will aid in instituting appropriate interventions to halt the progression of these complications. In cases where neuropathy was detected, participants were advised to seek further evaluation and management at the nearest health centre. Additionally, they were provided with individual health education on proper foot care practices.