The prevalence of painful diabetic neuropathy in 300 Moroccan diabetics

Painful diabetic neuropathy is a frequent complication of diabetes. Its diagnosis is clinical. Our goal is to determine the prevalence of painful diabetic neuropathy in this population. We also analyzed the relationship between this neuropathy and certain parameters, concerning the patient and his diabetes. It is a cross sectional study conducted at the department of endocrinology and internal medicine of Avicenne hospital Marrakech-Morocco, among a cohort of 300 diabetic outpatients. We used the DN4 questionnaire (Douleur Neuropathique en 4 questions), for diagnosis. The results showed a prevalence of 15%. In this study: advanced age, female gender, duration of diabetes greater than 10 years, and the lack of medical follow up were found to be statistically significant risk factors for painful diabetic neuropathy, in addition to some diabetes-related comorbidities such as hypertension, dyslipidemia, sedentary life style and diabetic retinopathy. Painful diabetic neuropathy remains undertreated, in fact 74% of our patients did not receive any specific treatment, knowing that the progress in developing effective and well-tolerated therapies has been disappointing.


Introduction
Diabetes mellitus is one of the biggest global health problems of the 21 st century. Its global prevalence in adults aged 20-79 years was estimated to be 8.8% [1]. In Morocco, it was estimated to be 7.7% [1]. Diabetes could lead to serious complications; in this study we were interested in the painful form of peripheral neuropathy. The prevalence of painful diabetic neuropathy (PDN) varies between 10% and 60% [2]. Patients should be systematically questioned concerning suggestive symptoms as they do not spontaneously report them. PDN can have debilitating consequences with significant impact on the quality of life and cost of care. It is a complication that affects small fibers, thus it may be accompanied by a normal monofilament test and electromyogram (EMG) [3]. The diagnosis is clinical, based on clinical finding, and has been facilitated by the development of simple and validated questionnaires, such as the neuropathic pain 4 questions (DN4) [4]. To date, glycemic control remains the most effective way to slow down and prevent PDN. In this study, we have determined the prevalence of PDN and its potential risk factors.

Methods
It is an observational, cross-sectional study, including 300 Moroccan diabetic out-patients, recruited from the departments of internal medicine and endocrinology departments, in Avicenne Hospital in Marrakech. A score of ≥ 4 on the DN4 was used to establish the diagnosis of PDN (Annex 1). The exclusion criteria were as follows: duration of type 1 diabetes mellitus less than 5 years, the presence of a psychiatric disorder that could influence the reliability of the questionnaire, patients with lower limb amputation, the presence of any other type cause of neuropathic pain (such as: post-herpetic neuralgia, pain associated with cancer, pain related to spinal cord injury, multiple sclerosis), being under certain drugs witch can give peripheral neuropathy (antiretroviral therapy , cisplatin, thalidomide, metronidazole). The statistical analysis was performed using SPSS V.20. Variable were compared using the χ 2 and Fisher exact tests and p-value <0.05 was con-sidered to be statistically significant.

Results
The mean age of patients enrolled in the study was 57.24 ± 9.79 years [32-85 years]. Males were predominant with a sex ratio (F/M) = 0.93. The men body mass index (BMI) was 27.53 ± 3.67 Kg / m 2 , 59.3% of these patients were overweight and 18% were obese ( Figure 1). The majority of the patients had Type 2 diabetes (95.7%), with a mean duration of 10.63 ± 7.47 years. Also 95.7% of the patients were followed-up for their diabetes. The mean glycosylated hemoglobin (HbA1c) of the last 3 months was 7.29% ± 1.37% ( Figure   2). The treatment received by the patients for diabetes manage-ment is summarized on Figure 3. A history of coronary artery disease (CAD) was found in 11% of patients, followed by peripheral artery disease (PAD) and ischemic stroke in 2.7% of cases each. In this cohort, 15.4% (95% CI [11][12][13][14][15][16][17][18][19] of the patients had a DN4 score ≥ 4 and met the diagnostic criteria of PDN. The mean DN4 score was 1.37 ± 1.98 ( Figure 4). Burning (23.7%), and tingling (23.7%) were the most common symptoms and were reported equally, followed by numbness (23.3%) and electric shocks (15.7%).
In the study of the relationship between PDN and certain parameters, concerning the patient and his diabetes, advanced age was significantly associated with the presence of PDN (p = 0.017). In fact, a linear relationship was observed between PDN and increasing age (Table 1). Also, female gender (p=0.03), duration of diabetes >10 years (p = 0.019) and the absence of diabetes follow-up (p<0.0001) were all associated with the presence of PDN (Table 2). Although obesity (BMI>30) and poor glycemic control (HbA1c >7%) were found to be associated with PDN, the data were not statistically significant. Almost all of our patients had type 2 diabetes, as such, a relationship between the type of diabetes and PDN couldn't be established in this study. Cardiovascular risk factors that were found to be significantly associated with PDN are: hypertension (p=0.001), dyslipidemia (p <0.0001) and sedentary life style (p <0.0001).
Chronic alcoholism and smoking didn't show any association. For degenerative complications, retinopathy was found to have a strong association with the presence of PDN (p <0.001). While nephropathy, PAD, CAD and a history of ischemic stroke showed no association to PDN in this study. In total, only 26% of our patients were treated for neuropathic pain, different medication used is presented on ( Table 3).

Discussion
Diabetic neuropathy is one of the most common complications of diabetes, affecting nearly half of diabetics [5], it is often asymptomatic. It is made of many syndromes; the most common form is Distal symmetric sensorimotor polyneuropathy (DSPN) [2].
Neuropathic pain is defined as pain arising as direct consequence of a lesion or disease affecting the somatosensory system [6]. The origin of pain in the PDN is not fully elucidated. Chronic hyperglycemia seems to play a key role in it, as it induces ionic dysregulation and hypoxia in peripheral neurons, particularly at the level of nociceptors [3]. It could also to excessive production of certain metabolites, such as methyl glyoxal which increases the excitability of nociceptive receptors [7]. Such hyperexcitability is also described at the level of the spinal ganglion, which constitutes a therapeutic target for pregabalin and gabapentin. Neuropathic pain can also occur acutely, following a rapid glycemic control, more often following an insulin therapy, referred to as insulin neuritis [3]. Other accessory mechanisms that could explain neuropathic pain is the role of sympathetic system in the sensitization and inhibition of pain [8]. This complexity of PDN pathophysiology offers a multitude levels of therapeutic interventions the future. Epidemiological data on PDN are rare, as few studies have focused solely on this form of diabetic neuropathy. The populations studied and the diagnostic tools are also highly variable. As a matter of fact, the prevalence of PDN is 8% in There is little data on risk factors and co-morbidities related to PDN.
In the literature, the link with advanced age is clearly established by several studies [10][11][12], including ours. In the present study the mean age of patients with neuropathic was 60.93 ± 9.76 years, moreover, we noted a statistically significant linear trend between increasing age and the prevalence of PDN. We found a significant higher prevalence of PDN in females compared to males with a F/M ratio close to 2, similar to other studies [10 , 12,13]. Diabetes duration is a major and well-established risk factor of DSPN, regardless of the patient's age.
A strong association have been reported between PDN and the duration of diabetes, especially after 10 years of evolution [10,11].
In the present study, 60.9% the patients PDN had diabetes duration >10 years. Hyperglycemia is another major risk factor of peripheral neuropathy [14], consequently a HbA1c> 6.5% increases its risk by 5 folds [15]. Therefore, a strict control of blood glucose might slow down or prevent the progression of PDN [16]. In the present study, the diagnosis of PDN was associated with a poor glycemic control: 63% of patients with PDN had a HbA1c> 7%, however, these findings were not statistically significant. Comparable to other studies, Hypertension was significantly associated with PDN, 70% of patients with a DN4 score >4 had hypertension [17,18]. Similarly, this study found a significant association between PDN and dyslipidemia, also demonstrated in other papers [19,20]. Sedentary life style was also found to be a risk factor of PDN in our study, similar results were reported by Sui-Whi et al [14]. Obesity has been documented as a risk factor for PDN [2,10,11]. Our study has noted a positive correlation between the two, however it wasn't statistically significant.  Table 4). The initial choice may also be influenced by the associated comorbidities (depression, insomnia). In case of therapeutic failure (improvement <30%) with the recommended maximum dose, then changing the drug class is recommended. If the efficacy is > 30%, and the pain remains > 3/10, it is recommended to add another therapeutic class. As a third line treatment tramadol or even morphine could be used. Other therapies are under study.

Conclusion
Despite its impact on quality of life and its high prevalence, PDN remains largely under-diagnosed and under-treated. The present study showed that almost 1 in 6 diabetics suffered from PDN, but only 26% received a specific treatment. The availability of a simple screening tool such as DN4 makes its detection and diagnosis easier.
Although there is a wide variety of therapeutic options, and many guidelines, none has proven to be satisfactory. Thus, more effective therapeutic combination should be found with limited side effects.

Competing interests
The authors declare no competing interests.

Authors' contributions
Hicham Baizri and Bouchra Lahmar: realized the design of the study,