PET in Epilepsy Surgery Planning: Finding Where Seizures Begin

For many people with epilepsy, seizures are controlled by medicine. For others they are not, and a question follows: could the seizures be coming from one small, well-defined place that might be treated surgically? Answering it takes a lot of evidence from several kinds of test, and PET is sometimes one of them. This page explains what an FDG-PET scan between seizures shows, how an epilepsy team uses it alongside EEG and MRI, and what it cannot do. It is general education, and the people treating you or your child decide which tests are right.

When surgery comes into the conversation

A paediatric neurologist interviewed by Mayo Clinic notes that about a third of people with epilepsy continue to have seizures despite appropriate treatment, and that for these people an evaluation for surgery can be an option. Mayo's patient information adds that surgery is usually done when tests show that seizures start in a small, well-defined area of the brain and that removing it would not affect vital functions such as speech, language, movement, vision or hearing.

That is why so many tests are involved. Before surgery can even be discussed, the team needs to know where the seizures begin, and they need independent tests to agree. A neurologist who specialises in epilepsy, often called an epileptologist, usually leads this work, with neurosurgeons, neuropsychologists and imaging specialists. If you are new to the reports that come out of such tests, reading a PET report explains some of the vocabulary.

What an interictal FDG-PET scan shows

"Interictal" means between seizures. The tracer, FDG, behaves like glucose, so the scan maps how much sugar each part of the brain is using at the time (the tracer is introduced in PET radiotracers explained). In many people with focal epilepsy, the region where seizures start uses less glucose between seizures than the tissue around it. A review in Current Neurology and Neuroscience Reports describes how, after FDG was developed in the 1980s, it was soon recognised that focal interictal hypometabolism correlated with the location of the epileptic focus, and the method came into wide use in presurgical evaluation. Mayo's patient information puts it plainly: areas of the brain with low metabolism may indicate places where seizures occur.

Reduced metabolism is a clue about location. It is not a picture of the seizures themselves. Nothing in an interictal scan shows a seizure happening, which is one reason teams compare it with tests that do, such as video EEG, which records brain electrical activity and the person's behaviour together.

How PET fits with EEG, MRI and other tests

The Mayo patient pages list the usual set of tests: EEG, MRI, sometimes functional MRI to map speech and movement areas before surgery, PET, SPECT (a related scan that maps blood flow during a seizure), neuropsychological testing and magnetoencephalography, which measures the magnetic fields of brain activity. PET is one item in that list, rather than the centrepiece. MRI looks at structure, and it is where many surgical targets are first found. EEG shows electrical timing. PET adds a third kind of information, metabolism.

The most useful moments for PET tend to be when the other tests do not give a clear answer. In a 2001 study in the Journal of Nuclear Medicine, 55 adults with hard-to-treat partial epilepsy had interictal FDG-PET, and the scan was read as localising in 76 percent of them: 90 percent of those with temporal lobe epilepsy and 36 percent with epilepsy outside the temporal lobe. The scan localised the focus in about 69 percent of people whose MRI had not localised it. When PET and MRI or EEG each pointed to a site, the sites matched in every case. A smaller retrospective review of a paediatric epilepsy programme, summarised by the American Epilepsy Society, described three children with a negative MRI but a positive PET; two of them had surgery after further monitoring, the removed tissue showed a developmental abnormality of the cortex called cortical dysplasia, and both remained seizure free at the time of the report. Both studies were small and from single centres, so they illustrate the idea more than they settle it.

How PET informs the surgical decision

At a general level, a team asks whether the evidence from every test points to the same place. When PET agrees with EEG and MRI, that agreement strengthens the case that the region is the true source and can be treated. When PET shows reduced metabolism somewhere else, the team has a reason to pause, repeat or extend testing. In the 2001 study, 21 of 24 people who went on to surgery had a localising PET, and 18 of those 21 (86 percent) were seizure free afterwards. The authors cautioned that the group was selected and small, and the difference from those whose PET did not localise was not statistically significant.

PET can also guide the next step. A 2022 paper in Radiologia Brasileira notes that the area of reduced metabolism may be larger than the lesion seen on MRI. It also says hypometabolism typically extends to areas surrounding the epileptogenic zone, the tissue that generates seizures. This is why PET is generally treated as a pointer toward a region, with the exact limits of any surgery decided using other information. Surgical decisions themselves rest with the epilepsy team and the patient, informed by every test, and also by the risks, which Mayo notes can in a small number of people include a permanent change in thinking abilities.

Limits worth knowing

Sensitivity varies by where the seizures start. The 2022 review reports that, depending on scanner resolution, interictal FDG-PET detects hypometabolism in nearly 80 percent of temporal lobe epilepsy, but in only 52 percent of frontal lobe epilepsy and about 53 percent overall. Seizures that start outside the temporal lobe are harder to localise, as the 2001 study also found.

The role of PET has also shifted over time. The Current Neurology and Neuroscience Reports review notes that the relevance of FDG-PET has slowly diminished since high-resolution MRI became available, although FDG remains the workhorse tracer in clinical PET. It describes newer tracers aimed at specific receptors or processes, but these have been tested only in small groups, and one that looked promising has a half-life of about 20 minutes and is produced at just a few centres worldwide. In other words, the tracers that appear in research papers are often not what is available at a given hospital.

Finally, reading these scans takes experience. Normal-looking scans do not rule out epilepsy, and abnormal-looking ones do not prove which area is responsible. For how PET and MRI can be combined in one session, see PET-CT and PET-MRI; for general preparation, what to expect on scan day explains the routine, including what to ask the imaging team about preparation, since individual protocols differ by centre. The wider family of brain applications is introduced in PET in dementia and memory evaluation.

The short version

Interictal FDG-PET shows where brain metabolism is reduced between seizures, which often overlaps with the area where seizures begin. Epilepsy teams use it as one clue among several, most helpfully when MRI and EEG do not give a clear answer. It is less informative for seizures starting outside the temporal lobe, and the decision about surgery rests on all of the evidence together.