Evaluation of serum high-mobility group box 1 concentration in dogs with epilepsy: A case-control study

Abstract

Background: High-mobility group box 1 (HMGB1) is a key mediator of neuroinflammation and there are increased HMGB1 levels in laboratory animal models of epilepsy and human patients with epilepsy.

Objectives: To determine serum HMGB1 levels in dogs with epilepsy.

Animals: Twenty-eight epileptic dogs, 12 dogs with nonepileptic brain diseases, and 26 healthy dogs.

Methods: In this case-control study, serum HMGB1 concentrations were estimated using the canine-specific enzyme-linked immunosorbent assay kit. Diagnosis of dogs with epilepsy was based on medical history, physical and neurological examination findings, laboratory test results, magnetic resonance image, and cerebrospinal fluid analysis.

Results: Serum HMGB1 levels were significantly higher in epileptic dogs (median = 0.41 ng/mL; range, 0.03-5.28) than in healthy dogs (median = 0.12 ng/mL; range, 0.02-1.45; P = .002). In contrast, serum HMGB1 levels of dogs with non-epileptic brain diseases (median = 0.19 ng/mL; range, 0.03-1.04) were not significantly increased compared to those of healthy dogs (P = .12). Regarding idiopathic epilepsy, dogs with an epilepsy course of >3 months showed a higher serum HMGB1 concentration (median = 0.87 ng/mL; range, 0.42-2.88) than those with that of ≤3 months (median = 0.26 ng/mL; range, 0.03-0.88; P = .02).

Conclusions and clinical importance: Serum HMGB1 could be a biomarker of epilepsy.

Keywords: HMGB1; biomarker; canine; seizure.

FIGURE 1

A, Serum HMGB1 concentrations of healthy control dogs compared to those of dogs with epilepsy and nonepileptic brain disease. B, Serum HMGB1 concentrations of healthy control dogs compared to those of dogs with idiopathic epilepsy, structural epilepsy, and nonepileptic brain disease. C, Serum HMGB1 concentrations of healthy control dogs compared to those of dogs with epileptic meningoencephalitis of unknown origin (epileptic MUO) and epileptic brain tumor (epileptic BT). The horizontal lines indicate the medians and interquartile ranges. The asterisk indicates statistically significant difference after Mann–Whitney U test (**P < .01, *P < .05)

FIGURE 2

Receiver operating characteristic curve illustrating the sensitivity and specificity for the use of HMGB1 to distinguish dogs with epilepsy from those without epilepsy. The line marked by triangles represents the area under curve that is 50%. The area under the curve of the receiver operating characteristic curve line marked by rectangle symbols is 0.75 (95% confidence interval = 0.6134‐0.8769). The point of intersection indicates the optimal cutoff of 0.15 ng/mL for the differentiation between dogs with epilepsy and without epilepsy, with corresponding sensitivity and specificity of 82.1% (63.11%‐93.94%) and 65.4% (44.33%‐82.78%), respectively

FIGURE 3

Comparison of serum HMGB1 concentration between before and after the treatment of epilepsy and underlying brain disease in epilepsy (P = .14; A), idiopathic epilepsy (P = .05; B), and structural epilepsy (P = .70; C)

FIGURE 4

Correlation between delta and sample interval in dogs with epilepsy (P = .54, r = 0.15; A), idiopathic epilepsy (IE) (P = .72, r = −0.13; B), and structural epilepsy (SE) (P = .61, r = 0.19; C). Correlation between ratio and sample interval in dogs with epilepsy (P = .65, r = 0.11; D), IE (P = .24, r = −0.43; E), and SE (P = .61, r = 0.19; F)

FIGURE 5

A, Correlation between serum HMGB1 concentrations and brain lesion volume (P = .42, r = −0.25) in dogs with structural epilepsy. B, Correlation between serum HMGB1 concentrations and brain lesion volume (P = .45, r = −0.2) in dogs with nonepileptic brain disease