Clinical utility of the DN4 questionnaire in the assessment of neuropathic pain in patients with type 2 diabetes: experience from a newly-created diabetes clinic in Cameroon

¹District Hospital Bafang, Ministry of Public Health, Bafang, Cameroon, ²District Hospital Nguelemendouka, Ministry of Public Health, Nguelemendouka, Cameroon, ³National Obesity Centre and Endocrinology and Metabolic Diseases Unit, Yaoundé Central Hospital, Yaoundé, Cameroon, ⁴Department of Public Health, Faculty of Medicine and Biomedical Sciences, University of Yaoundé 1, Yaoundé, Cameroon

^&Corresponding author
Jean-Claude Katte, Department of Public Health, Faculty of Medicine and Biomedical Sciences, University of Yaoundé 1, Yaoundé, Cameroon

Introduction    

Diabetic peripheral neuropathy is the most common microvascular complication of diabetes mellitus. A widely accepted definition of diabetic peripheral neuropathy is “the presence of symptoms and/or signs of peripheral nerve dysfunction in people with diabetes after exclusion of other causes” [1]. Neuropathies severely decrease patients' quality of life, as its sensory and motor symptoms can be very unpleasant. Also, unchecked neuropathy increases the risk of secondary complications such as foot ulcers, and eventual amputation, making early diagnosis very important [2]. Various models, scores and diagnostic kits are available to assess, diagnose and grade the severity of diabetic sensory neuropathy [3]. These include; The United Kingdom Screening Test, “Douleur neuropathique en 4 questions”, clinical stratification method using the Toronto Clinical Neuropathy Score, the Neuropathy Total Symptom Score (NTSS-6), the 10g Semmes-Weinstein Monofilament Test, Intra-epidermal nerve fiber analysis of skin biopsy, and nerve conduction studies. Although nerve conduction studies remain the most reliable, accurate and sensitive measure of peripheral nerve functions, vibration perception abnormalities have been shown to be the earliest finding in peripheral neuropathy [4]. Adequate nerve conduction and vibration perception studies are often not easy to perform in clinical settings because they are expensive and time-consuming. Simple screening tools such as the “Douleur neuropathique en 4 questions” form (DN4 questionnaire) may therefore offer a simple, practical alternative especially in resource depleted settings. The DN4 Questionnaire is a screening tool for neuropathic pain consisting of 10 interview questions (DN4-interview) and physical tests which has been validated for the diagnosis of diabetic peripheral neuropathy in western populations [5]. We therefore aimed to assess the performance of the DN4 questionnaire in detecting neuropathic pain associated with diabetic peripheral neuropathy in patients with type 2 diabetes with lower limb painful symptoms, compared to the more accurate vibration perception threshold determined by a digital biothesiometer.

Methods     

Design and setting: this was a descriptive and analytic survey carried out at the newly-created diabetes clinic (diabetes and hypertension treatment centre) of the Bafoussam Regional Clinic from June to December 2015. This diabetes clinic started in March 2013 as a community-hospital participatory effort. The Bafoussam Regional Hospital is a regional referral hospital that serves 22 district hospitals in the west region of Cameroon.

Ethical consideration and confidentiality: administrative authorization to conduct the survey was obtained from the Bafoussam Regional Hospital. All diabetes patients involved provided a written informed consent prior to the start of the survey. All patient data were collected with anonymized questionnaires.

Study participants: a total of 100 adult type 2 diabetes patients follow-up at the newly-created diabetes clinic were randomly selected from their medical records to participate in the survey. Fifty-eight (58) patients were finally included to undergo the survey based on complaints of paresthesia and pain at their lower limbs for a minimum period of 3 months prior to the survey date irrespective of gender. Figure 1 shows the flow diagram for the study.

Data collection tools: socio-demographic characteristics and medical history were recorded onto a pre-structured questionnaire. Biophysical characteristics were measured and recorded after a complete medical examination. The presence of neuropathic pain was assessed using the DN4 questionnaire. Fasting capillary glucose was measured using a glucose meter (Accu-Chek® Active, Roche, Germany).

Assessment for neuropathic pain using the DN4 Questionnaire: the DN4 questionnaire was used to screen for the presence of neuropathic pain. This questionnaire consisted of 4 sections; 3 sections concerned with symptoms review and associated symptoms and the 4^th section reserved for physical examination. DN4-interview questions included burning, painful cold, electric shocks, tingling, pins and needles, numbness and itching. Examination of the patient included hypoesthesia to touch, hypoesthesia to prick, and brushing. This was evaluated using the owen mumford® Neuropen. Each item was scored “yes” or “no.” Total questions across the 4 sections summed up to 10, with “yes” answer scored as 1, and “no“ as 0. Neuropathic pain was diagnosed if the total DN4 score was greater than or equal to 4 [5].

Assessment of vibration perception threshold (VPT): a single investigator (blinded to the DN4 Questionnaire) used a digital biothesiometer (VibroSense® - Genesis Medical Systems PvT. LtD, Telangana, India) to measure the vibration perception threshold (VPT) following a standardized procedure. The vibrating Biothesiometer probe was applied at an angle perpendicular to the testing site. The VPT was the average of 6 point measurement at the plantar surface of the big toe, 1^st metatarsal, 3^rd metatarsal, and 5^th metatarsal, instep, and heel. Once the probe was placed on the testing site, the voltage was slowly increased at the rate of 1 mV/sec and the VPT value was defined as the voltage level when the patient indicated that he/she felt the vibration. This was measured in both feet and the highest score was noted as the VPT for that particular patient. VPT was graded as Normal ≤15 V, Grade I :16-25 V, and Grade II > 25 V [6].

Statistical analysis: all data recorded onto the pre-structured questionnaires were entered into Microsoft Excel 2010 spreadsheet. All entries on computer were further checked against that on paper item by item to ensure accuracy. Data set was later exported to IBM SPSS 23.0 for analysis. Results were presented as frequencies or mean and standard deviation. The student´s t-test and the non-parametric equivalent were used to make comparisons where necessary. A p value of less than 0.05 was considered statistically significant for all analyses.

Results     

Socio-demographic characteristics of the study population: of the 58 diabetes patients screened for neuropathic pain, 31 (54.5%) were males. The mean age of the study population was 62.1 ± 9.9 years, with a median duration of diabetes of 3.5 (1.0-11.5) years and a median fasting blood glucose of 130 (94.0-168.5) mg/dl. The mean weight of the participants was 80.1 ± 18.1 kg with a mean body mass index (BMI) of 29.1 ± 5.8 kg/m². Overall 35 (60%) participants had hypertension with 54 subjects having a positive psycho-stimulant (alcohol and tobacco) consumption history as shown in Table 1.

Assessment of diabetic peripheral neuropathy based on vibration perception threshold: fifty-two (52) participants underwent the assessment for vibration perception threshold. A total of 43 people had neuropathy according to VPT findings. The mean VPT score was higher in patients with DN4 score ≥ 4 compared to those with VPT score < 4; 21.7 ± 9.0V vs. 26.3 ± 10.2V, p=0.108. Table 2

Frequency of neuropathic pain and the diagnostic accuracy of the DN4 Questionnaire: neuropathic pain was determined in 33 (56.9%) of the 58 participants with a DN4 score ≥ 4. Patients with DN4 score ≥ 4 were older than those with DN4 < 4, 63.8 ± 9.5 years vs 59.8 ± 10.3 years, p=0.125. Table 3 shows the characteristics of the study participants with regard to their DN4 scores. The sensitivity and specificity of the DN4 was 67.4% and 66.7% respectively with a positive predictive value (PPV) of 90%, and a negative predictive value of 70.0%.

Discussion     

We aimed to investigate the performance of the DN4 Questionnaire in assessing neuropathic pain in patients with type 2 diabetes with peripheral neuropathy diagnosed by vibration perception thresholds. We found that 56.9% of our study participants had neuropathic pain by the DN4 Questionnaire. The DN4 questionnaire however had a low sensitivity and specificity but with a high positive predictive value. Its high positive predictive value is comforting to the clinician as it suggests that a positive diagnosis reliably points to the presence of neuropathy. The PPV of 90.6% is high, and comparable to the findings of Timmerman et al. [7]. The sensitivity of the DN4 found in our study was 64.7%. This is lower than the 79.8% found by Perez in Spain [8] and can be explained by the low duration of diabetes in our study population with a median duration of just 3.5 years. The development of microvascular complications in diabetes increases exponentially with increasing duration of diabetes [9]. As such, the low sensitivity found here is not alarming, when compared to their duration of diabetes. Also worth noting that 43% of our study population attained primary education only. This might have influenced the sensitivity of the DN4 interview questions, as participant intellect and comprehension of the questions asked is an important factor in boosting accuracy. This study however has some limitations. The small sample size may have affected the overall power to show a difference between the DN4 score categories. This also did not allow us to examine for characteristics associated with neuropathic pain as other studies have reported [10]. Nevertheless, this is one of the first studies investigating the performance of a simple tool such as the DN4 Questionnaire in assessing neuropathic pain in a diabetic population in sub-Saharan Africa. Both the administration of the DN4 Questionnaire and the use of the digital biothesiometer require appropriate training which may booster the performance of such tools. Also, as peripheral neuropathy is a progressive continuum of symptoms, with vibration abnormalities first, followed by pain and other sensory abnormalities, lack of electromyography meant that accurate diagnosis of peripheral neuropathy type/stage was impossible. As such, the DN4 questionnaire would have missed early neuropathy cases not yet manifesting as pain which in some cases may have been picked by the vibration perception threshold.

Conclusion     

The DN4 questionnaire showed low sensitivity and specificity but with a high positive predictive value in identifying diabetic peripheral neuropathy. Therefore, the DN4 questionnaire may serve as an adequate tool in resource-limited settings given its simplistic and pragmatic approach, but would need adequate training in the administration of the interview questions and extensive validation in a larger population of patients with type 2 diabetes.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

JCK, designed and collected the data for the study. DT and JCK performed the statistical analysis. BBA, DT, AOB, and JCK wrote and reviewed the manuscript. All authors read and approved the final manuscript.

Acknowledgments     

All the staffs of the diabetes and hypertension treatment centre of the Bafoussam Regional Hospital who contributed to the realization of this patient survey.

Tables and figure     

Table 1: general characteristics of the study population

Table 2: categories by assessment of DN4 questionnaire

Table 3: diagnostic accuracy of DN4 questionnaire score compared to VPT

Figure 1: flow diagram of participant recruitment in the study

References