Serum Insulin and Leptin Levels in Children with Epilepsy on Valproate-associated Obesity.

BACKGROUND: Weight gain is a common adverse effect of sodium valproic acid (VPA) in children with epilepsy. Several mechanisms of VPA-induced obesity have been suggested such as increased appetite, facultative thermogenesis, and elevated insulin and leptin levels. In this study, we aimed to investigate the role of Insulin and Leptin in the pathogenesis of weight gain caused by VPA.
MATERIALS AND METHODS: Body mass index (BMI) was calculated, and serum insulin and leptin levels were measured in 45 consecutive patients and 45 controls.
RESULTS: The mean BMI of the cases and control group was 22.97 kg/m2 and 19.4 kg/m2, respectively, and it was significantly higher in cases (P < 0.001). Fasting serum insulin levels were higher in VPA group (26.3 μU/ml) than in controls (15.83 μU/ml), which was statistically significant (P < 0.001). Serum leptin levels were also found to be elevated significantly in VPA group (7.9 ng/ml) than in controls (1.6 ng/ml).
CONCLUSION: Sodium VPA is associated with significant rise of BMI, hyperinsulinemia, raised insulin resistance, and increased leptin levels in children with epilepsy.

INTRODUCTION

Epilepsy is a common and diverse set of chronic neurological disorders characterized by recurrent episodes of seizure. The mainstay of epilepsy treatment is long-term antiepileptic drugs. Sodium valproate (VPA) is an extensively used broad-spectrum antiepileptic drug. Although it is regarded as safe, use of VPA is associated with common side effects such as nausea, vomiting, hair loss, and rarely hepatotoxicity.[1] Among other side effects, weight gain is the frequently reported event although exact magnitude of the problem is unknown. Exact pathogenesis is still unknown, but it had been postulated that VPA-related weight gain is associated with increased insulin and leptin levels.[23] Insulin is an anabolic hormone that promotes glucose uptake, glycogenesis, lipogenesis, and protein synthesis of skeletal muscle. It is likely that leptin regulates body weight through Neuropeptide Y, which stimulates food intake and decreases thermogenesis in hypothalamus.[4] There is a paucity of literature in establishing the association of insulin level with that of VPA therapy. The purpose of this study was to evaluate possible changes in insulin and leptin levels during VPA therapy and their association with obesity.

MATERIALS AND METHODS

The study was carried out in the Department of Pediatrics at Postgraduate Institute of Medical Education and Research and Ram Manohar Lohia Hospital, New Delhi, from August 2014 to November 2015, with the approval of Hospital Ethical Committee. Written informed consent from parents of patients and healthy controls was obtained.

Study design

Observational cross-section study.

Study subjects

Forty-five consecutive patients aged 2–16 years were enrolled in the study group. Only those patients were included who were on VPA monotherapy for at least 6 months duration. Patients diagnosed with diabetes and metabolic syndromes were excluded from the study group. Epilepsy was classified according to the recommendation of International League against Epilepsy. Forty-five children of patient siblings served as healthy controls.

Study methods

The medical history and examination were taken on outpatient basis in a fixed pro forma. On clinical examination, weight, height, and Body mass index (BMI) were assessed. Body weight was determined to the nearest 0.1 kg and height was measured in triplicate with Harpenden stadiometer to the nearest 0.1 cm. Patients with BMI >95th percentile were considered obese and patients with BMI between 85th and 95th percentile were taken as overweight. Blood samples for serum insulin, leptin, serum glucose, and VPA drug concentration were obtained after overnight fasting. Insulin resistance was calculated using homeostasis model assessment of insulin resistance (HOMA-IR) index.

Assays

Serum insulin concentrations were measured with an enzyme-linked immunosorbent assay. The sensitivity of the assay was 0.5 μU/ml. Fasting serum glucose was measured using glucose oxidase method. HOMA-IR was calculated as a product of fasting insulin and fasting glucose divided by 22.5.[5] Serum leptin levels were analyzed using two-step capture or sandwich immunoassay. The Emit 2000 Valproic Acid Assay which is a homogeneous enzyme immunoassay technique for quantitative assessment of VPA was used for measurement of drug concentration.

Statistical analysis

Data were analyzed using Student's t-test and Mann–Whitney U-test for uniform and unevenly distributed variables, respectively. Association between drug level and insulin level was analyzed by spearman correlation coefficient. All the data were analyzed using SSPS Inc. Version 17.0. Chicago software.

RESULTS

A total of 90 children were included in this study which comprised of 45 cases and 45 age- and sex-matched controls. The clinical profile of children is depicted in Table 1. The mean age of patients in case and control group was 9.85 ± 3.48 and 9.94 ± 3.49 years, respectively. The mean BMI of the cases and control group was 22.97 kg/m2 and 19.4 kg/m2, respectively, and it was significantly higher in patients who were taking VPA from at least 6 months (P < 0.001). A mean drug level of VPA was found to be in normal therapeutic range (63.12 ± 23.70).

Table 1

Clinical and biochemical characteristics of the study individuals

Fasting serum insulin levels were higher in VPA group (26.3 μU/ml) than controls (15.83 μU/ml), which was statistically significant (P < 0.001). Similarly, insulin resistance (HOMA index) was much higher in VPA group (5.66) than in control group (3.15). Serum leptin levels were also studied in both the groups and it was elevated significantly in VPA group (7.9 ng/ml) than controls (1.6 ng/ml). Significant difference of insulin and leptin levels in both the study groups is clearly obvious as shown in Figure 1 (box plot).

Figure 1

Box plot showing insulin and leptin levels in study individuals

Subgroup analysis was done among obese and normal patients who were on VPA as shown in Table 2. Insulin levels were significantly elevated in the obese group (48.35 ± 24.88) in comparison to those children who were having normal BMI (15.91 ± 13.51). The correlation coefficient between serum VPA levels with serum insulin (P = 0.72), insulin resistance and serum leptin (P = 0.36) levels shows no significant positive or negative correlation.

Table 2

Body mass index and insulin level in obese and normal subgroup of cases on valproate

DISCUSSION

Weight gain is one of the most common adverse effects with VPA, and it has been described in several studies.[12] The risk of developing weight gain and insulin resistance poses a great concern, due to the known strong association of this side effect with the incidence of metabolic syndrome and type 2 diabetes mellitus even in childhood.

Weight gain is more common in patients who were taking VPA. In a 1-year prospective study, Verrotti et al.[2] reported the obesity rate as 35% in 20 prepubertal girls with epilepsy. In the same study, hyperinsulinemia and insulin resistance was found to be seen in the overweight group. Egger and Brett[6] noted weight gain in 44 of the 100 children treated with VPA. The precise mechanisms underlying the VPA-associated weight gain, however, remains unclear. VPA-treated patients developed increased appetite, thirst, and quenching with calorie-rich beverages. All these behaviors indicate toward hypothalamic stimulation. More recently, a study carried out in Wistar strain female rat shows that treatment with VPA disrupts hypothalamic axis at the level of GnRH pulse generator.[7] Weight gain is also attributed to VPA-induced changes in insulin and leptin levels. Insulin level and insulin resistance (HOMA-IR) were significantly elevated in the treatment group in comparison to control in this study. VPA is a fatty acid derivative, and this increase in free long chain fatty acid dysregulates the action of insulin and promotes the onset of insulin resistance through the inhibition of various metabolic pathways such as glucose uptake, glycogenesis, and the glucose oxidation.[8] Serum insulin is also thought to be increased due to the inhibition of hepatic insulin metabolism by VPA. Increased insulin levels were also proposed to be due to the direct action of VPA on beta cells of the pancreas.[9]

Leptin, a hormone secreted by adipose tissue, acts primarily to regulate body weight and energy expenditure. Serum leptin levels were significantly raised in patients on VPA group in our study. Similar results were reported in studies done by Verrotti et al.[10] They demonstrated that after 1 year of VPA treatment, obese patient had significantly higher leptin levels as compared to controls. A study done by Pylvanen et al.[3] in contrast found no significant difference in leptin levels between patients treated with VPA and the control group. This study was of cross-section design; therefore, opportunity to know cause-effect relationship between weight gain and leptin levels is not possible.

CONCLUSION

Sodium VPA monotherapy may be associated with significant rise of BMI, hyperinsulinemia, and raised insulin resistance in children with epilepsy. These results suggest that weight gain during VPA therapy may be related to increased insulin levels and insulin resistance. The precise mechanism underlying VPA-associated weight gain still remains unclear, and further well-controlled studies are required to establish this side effect and its long-term complication. Although concerns about potential weight gain should be discussed with patients and their parents before therapy begins, these concerns should not preclude the use of VPA.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.