The cortical ETV entry site is 1 cm in front of the left/right coronal suture and 2.5-3 cm beside the midline, which is the non-eloquent area without important vessels. The surgical procedures followed the protocol in the present study.
Table 2
Comparative analysis of serum electrolytes before and after surgery
| | Epilepsy (N = 5) | No-epilepsy (N = 45) | P |
| Preoperative calcium | 2.59 (± 0.14) | 2.47 (± 0.18) | 0.1603 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative calcium | 2.19 (± 0.12) | 2.35 (± 0.14) | 0.0429* |
| Miss | 1 (20.00%) | 11(24.40%) | |
| Change of calcium | -0.44 (± 0.12) | 0.13 (± 0.17) | 0.0049* |
| Miss | 2 (40.00%) | 12 (26.70%) | |
| Preoperative phosphorus | 1.75 (± 0.38) | 1.59 (± 0.29) | 0.3903 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative phosphorus | 1.40 (± 0.50) | 1.35 (± 0.25) | 0.7571 |
| Miss | 1 (20.00%) | 11 (24.40%) | |
| Change of phosphorus | -0.41 (± 0.19) | -0.23 (± 0.27) | 0.1863 |
| Miss | 1 (20.00%) | 12 (26.70%) | |
| Preoperative albumin | 45.67 (± 2.16) | 43.68 (± 3.58) | 0.2230 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative albumin | 39.93 (± 2.38) | 40.14 (± 4.43) | 0.8601 |
| Miss | 1 (20.00%) | 13 (28.90%) | |
| Change of albumin | -6.52 (± 2.45) | -2.85 (± 4.58) | 0.0824 |
| Miss | 1 (20.00%) | 14 (31.10%) | |
| Preoperative blood glucose | 5.51 (± 0.85) | 4.94 (± 0.75) | 0.1653 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative blood glucose | 11.70 (± 7.01) | 5.48 (± 1.53) | 0.0916 |
| Mis | 1 (20.00%) | 9 (20.00%) | |
| Change of blood glucose | 6.26 (± 7.36) | 0.57 (± 1.43) | 0.2029 |
| Miss | 1 (20.00%) | 10 (22.20%) | |
| Preoperative sodium | 136.90 (± 4.46) | 139.00 (± 3.01) | 0.2396 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative sodium | 137.30 (± 6.63) | 139.00 (± 2.85) | 0.2979 |
| Miss | 1 (20.00%) | 8 (17.80%) | |
| Change of sodium | 0.35 (± 9.93) | 0.24 (± 3.51) | 0.8389 |
| Miss | 1 (20.00%) | 9 (20.00%) | |
| Preoperative chlorine | 102.40 (± 1.53) | 102.60 (± 3.20) | 0.7467 |
| Miss | 1 (20.00%) | 7 (15.60%) | |
| Postoperative chlorine | 96.25 (± 4.17) | 102.20 (± 4.02) | 0.0881 |
| Miss | 3 (60.00%) | 27 (60.00%) | |
| Change of chlorine | -2.90 (± -) | 0.53 (± 5.02) | 0.4747 |
| Miss | 4 (80.00%) | 29 (64.40%) | |
| Preoperative potassium | 4.47 (± 0.52) | 4.37 (± 0.48) | 0.6202 |
| Miss | 0 (0.00%) | 1 (2.20%) | |
| Postoperative potassium | 3.56 (± 0.38) | 4.12 (± 0.44) | 0.0250* |
| Miss | 1 (20.00%) | 8 (17.80%) | |
| Change of potassium | -0.97 (± 0.67) | -0.18 (± 0.54) | 0.0288* |
| Miss | 1 (20.00%) | 9 (20.00%) | |
| *p < 0.05 |
Miss indicates the proportion (%) of patients with missing values.
The serum calcium and potassium levels before and after surgery significantly differed with the occurrence of epilepsy (Table 2). Although the type of irrigation fluid and the occurrence of epilepsy could not be statistically analyzed, it was revealed that epilepsy occurred in children irrigated with Ringer’s solution, while no seizure was found with injection of normal saline.
All children with epilepsy received Ringer’s solution as irrigation fluid intraoperatively (Fig. 3), which might be related to changing the concentrations of calcium and potassium ions in CSF [9, 10, 11]. A previous study reported that Ca2+ level in CSF was higher in patients with neonatal-onset epilepsy [12], which was consistent with our finding. Another study demonstrated that perfusion of the inferior horn in chronic awake cats with a high potassium level in CSF revealed that K+ increased epileptogenic excitability [13]. This result also supported our speculation that Ringer’s solution is a compound sodium chloride solution, containing potassium and calcium, and potassium and calcium concentrations increased in CSF. Compared with older children, younger children may have insufficient calcium and potassium metabolism, and identification of these children is particularly important.
Our study showed that the changes in serum potassium and blood calcium levels were related to the occurrence of seizures. The possible mechanism was that surgery changes the permeability of blood-brain barrier (BBB), making some ions easier to pass through the BBB, and the imbalance of serum calcium and potassium levels leads to the imbalance of calcium and potassium levels in neurons, resulting in epilepsy [14, 15]. In addition, intraoperative administration of irrigation fluid can change the ion concentration in CSF, because there is an increase in the permeability of BBB, making some ions more likely to pass through the BBB and may lead to epilepsy.
The incidence of immediate postoperative epilepsy in this group of cases was 10%, it was 8.3% in children younger than 1 years old, followed by 40% in children aging 1–2 years old, and 22.7% in children younger than 2 years old (Fig. 4). Children younger than 2 years old in the present study were more likely to develop postoperative seizures and all received the Ringer’s solution. In particular, children who aged 1–2 years old had the highest rate of epilepsy, while children older than 2 years old did not have epilepsy. According to the literature, the incidence of epilepsy was higher in the youngest age-based group, with an estimate of 86 per 100,000 children per year in the first year of age, followed by a downward trend to about 23–31 per 100,000 people who aged 30–59 years old. In children, the incidence of epilepsy was the highest in the first year of life and declined to adults’ incidence at the age of 10 years old [16]. Another study reported that the incidence was markedly higher in the first year of life, and structural/metabolic etiologies were more likely to present at this age than in older individuals, which might explain the higher incidence [17]. The present study also reflected this age-related susceptibility phenomenon.
Status epilepticus (SE) is an epileptic seizure that is sufficiently prolonged or repeated at sufficiently brief intervals, so as to produce an enduring epileptic condition. SE may have long-term consequences, including neuronal injury or death and alteration of neuronal networks, depending on the type and duration of seizures [16]. In our study, 2 patients developed SE. Only strong sedative drugs can control the seizures. However, all patients showed frequent seizures within 6 h after surgery and seizures did not appear again after 24 h, which also indirectly reflected the promoting effect of the surgery itself on seizures, especially with the circulation of CSF, and irrigation fluid was partly removed from the CSF intraoperatively.
The outer diameter of the sheath of the rigid endoscope was 4 mm, while the outer diameter of the working channel of the flexible endoscope was 9 mm, which both slightly damaged to the cortex. The present study showed that the incidence of epilepsy caused by both a rigid endoscope and a flexible endoscope was not statistically significant.
In a child with no seizure after ETV, ventriculoperitoneal shunting was performed 3 weeks after the first surgery, and epilepsy occurred at 4 years of follow-up, which might be associated with shunt surgery, and a meta-analysis reported that the risk of acquiring seizures/epilepsy in shunted non-infectious hydrocephalus children was 15.75 times higher than that in normal children [18]. A randomized controlled trial compared the anterior and posterior shunt entry sites, and found no significant difference in the incidence rate of new-onset epilepsy [19], which could be related to cortical injury and long-term stimulation of the shunt as a foreign body, indicating that cortical disruption or irritation from the shunt catheter itself may contribute to post-shunt seizure development. The child has been well controlled by antiepileptic treatment.
Epileptic seizures are dangerous from a medical point of view, in which with increasing cerebral oxygen consumption, intracranial pressure and trauma could be enhanced [20]. At present, there are several antiepileptic drugs (AEDs), while few of them can be used for infants. Levetiracetam was previously recommended as a perioperative prophylactic antiepileptic medication, and there are also some researches on the administration of levetiracetam for infants with epilepsy [21–23]. Levetiracetam possesses an excellent oral bioavailability and a very low protein binding, with unknown significant pharmacokinetic interactions. There is no hepatic metabolism; 66% is excreted in urine as constant form and the rest is hydrolyzed as inactive compounds. Therefore, it is not affected by other drugs and does not influence the functions of liver and kidney. Levetiracetam is a broad-spectrum drug, which is effective against focal seizures, generalized tonic-clonic seizures, and generalized myoclonic seizures. Levetiracetam has demonstrated class I evidence of efficacy as an adjunctive therapy for refractory generalized myoclonic seizures, and is the only AED with FDA approval for treatment of such type of seizures [21, 24–25]. It is recommended for young children receiving a loading dose of levetiracetam (20 mg/kg) at the start of surgery and a maintenance dose (10 mg/kg twice per one day) for the following 7 days [8]. In a previous study, a patient’s age was found as an influential factor of the success rate of ETV, and younger patients tended to undergo additional surgery after failure of ETV [4]. As the reduction of postoperative serum potassium and calcium levels was found as a risk factor for epilepsy, the changes in serum calcium and serum potassium levels should be monitored before, during, and after surgery. For children with a downward trend of serum calcium and serum potassium levels, timely potassium and calcium supplementation during and after surgery may also reduce the incidence of epilepsy.
There are some limitations in the present study. Firstly, this was a retrospective study, and epilepsy was depended on documented seizures without routine postoperative detection using electroencephalogram. Secondly, due to the small sample size, the confounding factors were not excluded, which could limit the generalizability of our findings. Finally, no causal relationship between postoperative epilepsy and serum ion level was identified because the levels of potassium and calcium were not assessed before the epilepsy onset.