Why do patients with epilepsy need to be tested? This question often confuses patients and families, and sometimes even the treating physician does not understand the purpose of the test he is having the patient undergo, the appropriate timing, and the analysis of the results. Today, I came across three consecutive patients with different conditions that just brought out the problem. 1. Patient A, female, 30 years old. She was treated at an outside hospital for episodes of “unconsciousness and twitching of the extremities for 4 years”. She was treated with valproic acid, carbamazepine, lamotrigine and other drugs, but the effect was not good and she still had seizures once every 3-6 months. The EEG was also mildly abnormal after a review. Because of unsatisfactory treatment effect, he came to me today by name. I asked him about his medical history in detail: before each seizure, there were triggering factors such as bad mood or mental stimulation, and each seizure lasted 20 minutes to half an hour. Recently, there were family conflicts, which led to an increase in the frequency of seizures. Long-term medication but poorly controlled. After induction, the patient informed me that she had a family history of schizophrenia and that her uncle, aunt and cousin had all been diagnosed with “mental illness”. I suggested that she should have a long-range EEG. I explained to her that the diagnosis of “epilepsy” was suspicious from the onset of the disease and its treatment. It may not be epilepsy at all, but a so-called “pseudo” seizure, or it may be a combination of epilepsy and psychiatric illness, which would not be effective if only epilepsy is treated. Now we need to check the 24- to 48-hour EEG and re-evaluate the diagnosis before adjusting the treatment plan. 2. Patient B, female, 27 years old. Diagnosed with epilepsy at an outside hospital because of recurrent episodes of unconsciousness and convulsions for 6 years. The patient had each seizure at night during sleep and never during the day, so the treating physician at the initial consultation gave her carbamazepine 0.1 every night and told the patient that the dose was sufficient and should not be increased. After that, she was seizure-free for a year. Then she switched to topiramate because of pregnancy, but the seizures became more frequent, and switched back to carbamazepine again, with occasional mild upper extremity twitch-like seizures. In the last six months, the seizures started with perioral twitching and gradually developed into twitching of the extremities, which made it impossible to wake up, and last month, she had one “grand mal” seizure during nap time. A few days ago, she was referred to our hospital, where the treating doctor considered her to have “intractable epilepsy” and recommended that she be hospitalized for tests, including a 72-hour pterygoid electroencephalogram, cranial MRI, and even PET-CT, to evaluate the possibility of surgical treatment. The patient felt that the problem was serious, so she came to me through an acquaintance. I explained to her: from the seizure situation, it is focal epilepsy and only seizes at night, it is possible to take the medication at night and not during the day, which can be treated without any significant adverse effects. From the treatment effect, carbamazepine is indeed effective, as illustrated by the frequent seizures when switching to topiramate. Unfortunately, because the dose is too low to achieve the optimal effect of treatment, and has never been “seizure-free” for 6 years, it cannot really be diagnosed as “intractable epilepsy” for the time being, but at most considered as “medically induced” intractable epilepsy. The most that can be considered is “medical” refractory epilepsy. I suggested that the patient increase the carbamazepine dose to 0.2-0.3 per night. At the same time, I made her understand that epilepsy treatment is a long-term process and that if the medication is effective, surgery does not need to be considered. Surgery is only used for truly refractory epilepsy, and less than half of patients can completely discontinue their medication after surgery. Since most patients still need to take medication after surgery, surgery should not be performed in patients who are already controlled by medication, so surgery is only used as a supplement to failed medication rather than as a first option. At the same time, cranial surgery has certain risks, and once the surgery fails, the minor ones are ineffective, and the major ones are unable to speak, paralysis, personality change or even death. 3. Patient C, female, 8 years old. She was diagnosed with epilepsy in our hospital. She was treated with oxcarbazepine for 2 years without seizures and was considered “cured”. He was treated with oxcarbazepine for 2 years without seizures, and was considered “cured”. After the recurrence, the 24-hour EEG was twice rechecked without any abnormality, and the cranial MRI suspected “right temporal lobe atrophy”. The doctor recommended that the patient be hospitalized for pterygoid electrode EEG and PET-CT to evaluate the possibility of surgery. Today, he came to me to solve his problem. I explained to the family that the diagnosis of epilepsy was clear from the seizure symptoms and the previous EEG, and that the treatment was effective, with a reduction in seizures or even “no seizures”. The subsequent failure to reduce the medication requires an analysis of the reasons. (1) It is possible that the previous treatment did not achieve complete control: The family also recalled that although the girl was not found to have seizures during the day, sometimes her mouth would twitch and shout during sleep at night, or she would get up and walk around. The family always thought it was “sleeping sickness” and did not pay attention to it. I said that this might be a seizure, and that if we had not ruled it out in the first place, the seizure might not be a “relapse”, but a failure to cure. The subsequent EEG did not show any abnormal changes, so it was speculated that the seizure foci were deep and could not be detected because they were not transmitted to the surface of the brain due to the interference of drugs. (2) Possible progression of the disease resulting in increased seizures, commonly seen in diseases such as brain tumors. In such a case, the primary disease needs to be treated to control the epilepsy. The girl has recently undergone a cranial MRI, and no such disease was found. (3) Children are growing and developing, and as time goes by, the amount of medication will need to be increased to accommodate the increase in weight. The girl was able to control it before, but after the reduction of medication, it is estimated that the amount of medication is relatively insufficient, the treatment of this situation is simple, increase the dose of medication is good. The best point is that the child’s symptoms during the seizure show a left-sided lesion, and the EEG during the interictal period also shows a left-sided one, but the MRI shows an abnormality on the right side, which is a contradiction. To further define the epileptic lesion, a series of tests are needed with the ultimate goal of surgery! However, I think the patient does not need to consider surgery for the time being, and can observe the change of seizures after the medication is increased to the full amount first. With a clear diagnosis, it is not worthwhile to rush into a test with an unclear goal, risking both infection and interfering with the child’s studies.