T1DM is the most common subtype of pediatric diabetes mellitus, affecting about one million children and adolescents globally [7], the morbidity of T1DM is increasing worldwide, especially under 6-year-old children[8]. It is currently speculated that genetic and environmental factors play an important role in the pathogenesis of T1DM, and the latter factors include unreasonable eating habits, lack of physical activity, polluted environment, etc., all these complicate diabetes management and make it harder to maintain good metabolic control. The typical clinical manifestations of T1DM include polydipsia, polyuria, polydipsia and weight loss. Polyuria, increased nocturia, even enuresis, are often the onset symptoms, they should be considered the possibility of diabetes if they occurre in older children. Whereas the frequencies of diabetic ketoacidosis as one of the common complications range from approximately 15–70% in developed countries and diabetic ketoacidosis at onset is more common under 2-year-old children [9], therefore they are often complaint with mental malaise, blurred consciousness or even coma, prolonged and irregular breathing with smell of acetone, nausea, vomiting, abdominal pain, even shock. In the present study, 82.4% of children patients showed typical symptoms, and 25.9% of patients were admitted to hospital due to diabetic ketoacidosis in which 5.9% were the onset symptom.
As a major complication of diabetes, the morbidity of DPN varies from 9–97% [10], the common symptoms and signs include pain, numbness, foot ulceration and limb weakness. Compared to adults, children are more tend to have subclinical neuropathy (that is no signs or symptoms of DPN) in the early stage of the disease [11]. Subclinical DPN may be reversible under intensive interventions in the early stage, such as lifestyle changes and medical therapy, however, the cumulative nerve impairments eventually become irreversible and clinical signs or symptoms occur overtime, which seriously affect the treatment effect as well as prognosis of the children. Therefore, early detection of subclinical DPN in children diabetes might allow earlier intervention, then possibly reduce or delay the morbidity of clinical DPN in adult life.
Nerve conduction study (NCS) is mainly applied to assess the large myelinated motor and sensory nerves of limbs in the clinical [12], and is a simple, non-invasive and well-tolerated test to pick up early diagnosis and screening of DPN in children and adults [13, 14]. In the present study, we found that slowing of motor nerve conductive velocity mainly occurred in lower extremities, slowing of sensory nerve conductive velocity happened in upper and lower limbs, whereas the amplitudes of upper and lower limbs were no significances when compared to control subjects, suggesting that demyelination of large somatic nerve fibers were the main pathological changes of peripheral neuropathy and nerve fibers of the lower limbs were more frequently impaired, the abnormal rates of NCS were up to 48.2%,and which were earlier than clinical manifestation of peripheral nervous impairment in T1DM.
Autonomic dysfunction is an important factor of morbidity and mortality among pediatric DM [15, 16], the common evaluations include heart rate, blood pressure, dry skin, as well as the callus. However, some subjective questionnaires could not reflect any differences between the patients and healthy controls [17], whereas some objective function tests have not been applied for children owing to higher demand for cooperation or lack of specialized equipments. Autonomic neuropathy in children with T1DM is likely underestimated due to suboptimal screening and subclinical neuropathy [18]. Therefore, many studies have been shown that the SSR is a repeatability and non-invasive examination evaluating autonomic function in limbs [19, 20].The high abnormality of SSR in adult patients with diabetic neuropathy has been pointed out by many authors [21, 22]. SSR are made up of negative wave and positive wave, the parameters of SSR contain the latency (onset, N and P latency) and the amplitude. A lot of studies have shown that the amplitude of SSR was highly variable and unreliable parameter due to it was influenced by habituation and age, whereas the latency was a stable parameter, free from repeated stimuli [23, 24].In order to avoid the influence of age, we matched 30 age-matched healthy control subjects. In this study, we found that latencies of SSR were prolonged or absence in the upper and lower limbs, and the abnormal rates of SSR were up to 61.2%, especially occurred in the lower extremities, which were earlier than clinical manifestation of autonomic nervous impairment in T1DM,and may be a subclinical manifestation of T1DM. Meanwhile, the combination of SSR and NCS tests were superior to the single SSR or NCV test and provided the early accurate diagnosis of diabetic DPN.
There are many variable recommended guidelines at present, some experts advocate to start screening 5 years after the definitive diagnosis of T1DM in children [25], while some experts advocate to start screening 2–3 years of disease onset due to the rising incidence and complications of diabetes in children [26], therefore, SSR and NCS could play a prominent role in screening to develop DPN. In this study, we also analyzed the association between short or long duration of disease and the abnormal rates of SSR and NCV, and then found that the abnormal rates of SSR and NCS in the long duration of illness were higher than those in the short duration of illness, suggesting that the pediatric DPN were related to the duration of disease, which have been confirmed by most adult diabetes studies [27].Meanwhile, 65 patients with DPN had shown that the onset latencies of SSR were shortened and nerve conduction velocities were improved under intensive interventions for 6 months. This further proved the importance and necessity of SSR and NCS detection in diabetic children.
This study also had some limitations. Firstly, all enrolled pediatric diabetic patients were no signs or symptoms of peripheral neurology. we did not conduct cohort studies whether some patients with subclinical peripheral neurology would show obvious clinical manifestations in the future or not. Secondly, our study did not conduct other autonomic function tests except for SSR test in pediatric diabetes.
In conclusion, sympathetic skin response could evaluate the small unmyelinated sympathetic fibers and nerve conduction study could assess the large myelinated peripheral nerves, they were both generally well-tolerated by children and the combination of SSR and NCV tests were superior to the single SSR or NCV test, therefore, SSR and NCV tests could provide the accurate early diagnosis as well as follow-up of pediatric diabetic neuropathy.