Vasopressin in the cerebrospinal fluid of febrile children with or without seizures

Copeptin as a serum biomarker of febrile seizures

Background and Objectives

Accurate diagnosis of febrile seizures in children presenting after paroxysmal episodes associated with fever, is hampered by the lack of objective postictal biomarkers. The aim of our study was to investigate whether FS are associated with increased levels of serum copeptin, a robust marker of arginine vasopressin secretion.

Methods

This was a prospective emergency-setting cross-sectional study of 161 children between six months and five years of age. Of these, 83 were diagnosed with febrile seizures, 69 had a febrile infection without seizures and nine had epileptic seizures not triggered by infection. Serum copeptin and prolactin levels were measured in addition to standard clinical, neurophysiological, and laboratory assessment. Clinical Trial Registration: NCT01884766.

Results

Circulating copeptin was significantly higher in children with febrile seizures (median [interquartile range] 18.9 pmol/L [8.5-36.6]) compared to febrile controls (5.6 pmol/L [4.1-9.4]; p <0.001), with no differences between febrile and epileptic seizures (21.4 pmol/L [16.1-46.6]; p = 0.728). In a multivariable regression model, seizures were the major determinant of serum copeptin (beta 0.509; p <0.001), independently of clinical and baseline laboratory indices. The area under the receiver operating curve for copeptin was 0.824 (95% CI 0.753-0.881), significantly higher compared to prolactin (0.667 [0.585-0.742]; p <0.001). The diagnostic accuracy of copeptin increased with decreasing time elapsed since the convulsive event (at 120 min: 0.879 [0.806-0.932] and at <60 min: 0.975 [0.913-0.997]).

Conclusions

Circulating copeptin has high diagnostic accuracy in febrile seizures and may be a useful adjunct for accurately diagnosing postictal states in the emergency setting.

Diurnal variation in febrile convulsions

This study sought to determine diurnal variations in febrile convulsions, and to investigate whether such variations influenced the severity of febrile convulsions. The study involved 326 children, between ages 6 months and 6 years, with simple febrile convulsions. Data were collected systematically by interviewing witnesses within the week after febrile convulsions occurred. The frequency of febrile convulsions was approximately 5 times greater in the evening than in early morning. An adaptation curve revealed that the maximum occurrence of febrile convulsions was at 4:00 pm (4:00-4:59 pm), and the minimum, at 4:00 am (4:00-4:59 am), similar to the pattern of human body temperature. Temperature and seizure duration did not differ significantly between high-frequency and low-frequency zones (2:00-7:00 pm and 2:00-7:00 am, respectively) (high-frequency zone vs low-frequency zone, 39.20 degrees C (S.D., 0.68 degrees C) vs 39.22 degrees C (S.D., 0.64 degrees C) and 3.82 minutes (S.D., 5.27 minutes) vs 3.14 minutes (S.D., 3.19 minutes)). These results suggest that the circadian rhythm does not change seizure propensity, but its hourly occurrence is attributable to an elevation in the temperature set point in the 24-hour period. The prevention of recurrent febrile convulsions by rectal administration of anticonvulsants in high-frequency zones would be clinically helpful.

Blood-cerebrospinal fluid barrier in hyperthermia

The blood-CSF barrier (BCSFB) in choroid plexus works with the blood-brain barrier (BBB) in cerebral capillaries to stabilize the fluid environment of neurons. Dysfunction of either transport interface, i.e., BCSFB or BBB, causes augmented fluxes of ions, water and proteins into the CNS. These barrier disruptions lead to problems with edema and other compromised homeostatic mechanisms. Hyperthermic effects on BCSFB permeability and transport are not as well known as for BBB. However, it is becoming increasingly appreciated that elevated prostaglandin synthesis from fever/heat activation of cyclooxygenases (COXs) in the BCSFB promotes water and ion transfer from plasma to the ventricles; this harmful fluid movement into the CSF-brain interior can be attenuated by agents that inhibit the COXs. Moreover, new functional data from our laboratory animal model indicate that the BCSFB (choroidal epithelium) and the CSF-bordering ependymal cells are vulnerable to whole body hyperthermia (WBH). This is evidenced from the fact that rats subjected to 4h of heat stress (38 degrees C) showed a significant increase in the translocation of Evans blue and (131)Iodine from plasma to cisternal CSF, and manifested blue staining of the dorsal surface of the hippocampus and caudate nucleus. Degeneration of choroidal epithelial cells and underlying ependyma, a dilated ventricular space and damage to the underlying neuropil were frequent. A disrupted BCSFB is associated with a marked increase in edema formation in the hippocampus, caudate nucleus, thalamus and hypothalamus. Taken together, these findings suggest that the breaching of the BCSFB in hyperthermia significantly contributes to cell and tissue injuries in the CNS.

Circadian and seasonal variation of first febrile seizures

Time of occurrence of 188 first febrile seizures (FS) was recorded, both by four 6-hour periods and by hourly intervals. The frequency of events was significantly ( P < .001) increased from 6 to 11.59 pm with a peak between 5 and 8 pm . A seasonal peak was observed in January.

Lowering of the potassium concentration induces epileptiform activity in guinea-pig hippocampal slices

Extra- and intracellular recording techniques were used to study the epileptiform activity generated by guinea-pig hippocampal slices perfused with low potassium containing artificial cerebrospinal fluid. Extracellular field potentials were recorded in CA1 and CA3 regions along with intracellular recordings in CA3 subfield. Reduction of the extracellular potassium concentration [K(+)](o) from 4 to 2 mM caused a transient neuronal hyperpolarisation which was followed by a repolarisation and subsequent depolarisation period. Paroxysmal depolarisation shifts occurred during the transient hyperpolarisation period while epileptic field potentials (EFP) appeared in the late repolarisation or early depolarisation phase. EFP elicited by reduction of [K(+)](o) were neither affected by blockade of N-methyl-D-aspartate (NMDA) glutamate-subreceptor or gamma aminobutyric acid receptor, nor by application of the organic calcium channel blocker nifedipine or the anticonvulsant drugs carbamazepine and valproic acid. Upon application of non-NMDA glutamate-subreceptor blocker the EFP were abolished in all trials, while application of the organic calcium channel blocker verapamil only suppressed the EFP in some cases. The results point to a novel mechanism of epileptogenesis and may provide an in vitro model for the development of new drugs against difficult-to-treat epilepsy.