A systemic approach is paramount for interpreting an EEG recording. Before starting the analysis, certain factors including the patient's age, level of physical activity, mental state, level of consciousness, the influence of different biological factors, environmental factors, and pharmacological agents that can potentially influence the morphology of the waveforms need to be taken into consideration.

There is a wide variation in the EEG waveforms. A good understanding of the normal or benign variants is necessary to differential normal or benign variants from pathologic waveforms.[15] Some of these normal variants include:

1. Wicket spikes: These waveforms appear over the temporal (anterior or mid-temporal) region during relaxed wakefulness, drowsiness, or state of light sleep.
2. Benign epileptiform transients of sleep (BETS): These are also called as small sharp spikes (SSS) and occur in stage 1 or stage 2 of sleep.
3. 6 Hz "phantom" spike-and-wave complex (PhSW): PhSW can be considered a smaller version of the 3Hz spike and wave pattern. These waveforms have low amplitudes and appear in the frequency of 5 Hz to 7 Hz.
4. Rhythmic mid-temporal theta of drowsiness (psychomotor variant) or RMTD: This pattern is usually located in the mid temporal region and appears in relaxed wakefulness and drowsiness.
5. Positive occipital sharp transients of sleep (POSTS): These waveforms appear in the occipital regions during non-rapid eye movement (NREM) sleep and are asymmetrically distributed.
6. Subclinical rhythmic EEG discharge in adults (SREDA): This is a very rare benign EEG pattern. It resembles ictal discharges and is, at times, interpreted as such, leading to a misdiagnosis of epilepsy.
7. 14 Hz and 6 Hz positive spikes: 6 and 14 Hz positive spikes are typically seen in the younger age group.
8. Presence of repetitive vertex waves, especially in children.
9. Breach rhythm: These waveforms are seen over the regions that have a skull defect. Because of the skull defect, there is increased visibility of faster frequencies that are otherwise less appreciated on scalp EEG.

EEG recording of a seizure can begin with the appearance of abnormal discharges in bursts, known as ictal epileptiform discharges. The discharges increase in frequency to rapid continuous spikes and waves, progressing to numerous spikes with buried waves, at peak seizure activity. As the activity slows down, the waves reappear and progressively reduce in frequency, and finally stop.[16] The time of seizure activity is called the ictal period, and the time between seizures is called the interictal period. EEG activity during the interictal period also reveals abnormal discharges, and these are called interictal discharges (IEDs). Since most patients present either immediately after or before a seizure, IEDs are important for validating a clinical suspicion of seizure activity in epilepsy. 

Typically, multiple EEGs are required to record IEDs. To put things into perspective, every fourth consecutive EEG in an epileptic patient has an IED frequency between 60% to 90%. The frequency of IED in a non-epileptic patient is about 0.5% to 2.5% in healthy young men and about 12% in non-epileptic patients of all age groups with progressive cerebral disorders. Specificity is probably lower and sensitivity higher in children as compared to adults.[17]

Activation methods like hyperventilation, sleep deprivation, and photic stimulation increase the appearance of IEDs and are useful for localization purposes and in being more assertive about the diagnosis of epilepsy.

Although surface EEG recordings are less sensitive than invasive studies, they are efficient in approximating the epileptogenic zone in most of the common epilepsies. The most commonly used invasive electrodes are stereotactically implanted depth electrodes and subdural strip or grid electrodes. Invasive studies are most useful when the scalp EEG does not yield a result, or when the focus is located adjacent to the eloquent cortex.[18]

There are a few concerns regarding the waveforms in EEGs. Firstly, the choice of the reference electrode should be such that it cancels out the normal waveforms and amplifies only the pathological ones. It must be remembered that an EEG is influenced by all electrical activity occurring both remotely and locally; however, the closest electrical activity would be the most prominent influencer. The reference electrode should be placed in a location where it captures all the interfering waveforms as well. The next concern for the reference electrode is that there must be a significant potential difference to facilitate the charge movement without acceleration. This means that a reference electrode placed too close to the pathological site would essentially have the same potential as the active electrode, making the recording not useful for interpretation.[4] 

The most important waveforms for clinical evaluation include delta (0.5 Hz to 4 Hz); theta (4 Hz to 7 Hz); alpha (8 Hz to 12 Hz); sigma (12 Hz to 16 Hz) and beta (13 Hz to 30 Hz) waves. Additionally, other waveforms like infra slow oscillations (ISO) (less than 0.5 Hz) and high-frequency oscillations (HFOs) (greater than 30 Hz) have increasingly been considered clinically relevant with the discovery of digital signal processing.[19][20][21][22][23]

EEG is useful in a variety of cerebral pathologies and is complemented by a spectrum of more advanced imaging modalities. EGG has been used in the evaluation and study of epilepsy, states of altered consciousness, the parasomnias, dementias, toxic confusional states, cerebral infections, and various other encephalopathies. Abnormal waveforms in EEG reflect a wide spectrum of general pathophysiological processes, raised intracranial pressure, cerebral anoxia, edema, epileptogenesis, etc., and do not specify a particular disease most of the times.[24] EEG is fundamental in discerning an array of localized epileptiform activity in the cerebral cortex. 

There exists a dynamic interaction between specialized regions in the brain that enable processing of information. This interregional communication mediates an eclectic range of neurological processes. These intercortical communications are disrupted in patients with focal epilepsy.[25] However, clinical studies have recorded abnormal activity beyond the region of pathology.[26][27][28][29] 

In a study conducted by Foldvary et al., localized ictal onset was seen in 57% of seizures. These included mesial temporal lobe epilepsy (MTLE), lateral frontal lobe epilepsy (LFLE), and parietal lobe epilepsy. Lateralized onsets predominated in neocortical temporal lobe epilepsy, and generalized onsets were seen in mesial frontal lobe epilepsy (MFLE) and occipital lobe epilepsy.[30]

Classical EEG morphologies, based on specific lobular involvement and epileptic foci are described below.

Temporal Lobe Epilepsy (TLE)

1. Ictal EEG finding: Ictal EEG is abnormal in about 95% of the patients with focal impaired awareness seizures, with 66% of them showing an electro-decremental pattern. A 5 Hz to 7 Hz rhythmic theta discharge in the temporal regions is specific for temporal lobe epilepsy. Depth electrodes are useful in capturing this pattern and are accurate in diagnosing ipsilateral mesial temporal structures. Scalp electrodes successfully lateralize but fail to accurately localize the focus of the seizure. Ictal EEG changes are difficult to appreciate using the scalp electrodes in frontal lobe seizures because of movement artifacts. Most EEGs would require additional neuroimaging studies using various modalities like magnetic resonance imaging (MRI), interictal fluorodeoxyglucose (FDG) photon emission tomography (PET), ictal single-photon emission computed tomography (SPECT), magnetic encephalography (MEG) or functional magnetic resonance imaging (fMRI) for further assessment of the abnormalities.[31]
2. Interictal epileptiform discharges: Isolated IED like sharps and spikes over the temporal region and temporal intermittent rhythmic delta activity (TIRDA), in addition to clinical history, are strongly correlated with a diagnosis of temporal lobe epilepsy (TLE).[32][33] In medial temporal lobe seizures, the amplitude of mesial temporal spikes is maximal at the anterior temporal scalp electrodes, and sphenoidal electrodes when used.[34]
3. Seizure semiology:
   • Auras: Visceral sensations and fear are associated with mesial temporal lobe epilepsy, while auditory and vertiginous auras have been associated with the neocortical temporal lobe epilepsy.
   • Neocortical temporal lobe epilepsy: At seizure onset, patients with neocortical TLE are more likely to describe any type of hallucination or illusion. Automatisms and dystonic posturing are not found in neocortical TLE.[32][35]
   • Mesial temporal epilepsy: Patients present with a behavioral arrest that is observed as a blank facial expression along with loss of awareness. This change is followed by oral, facial, or alimentary automatisms such as lip-smacking, chewing, sucking, or swallowing or is accompanied by ipsilateral automatisms such as repetitive hand movements, picking, or fidgeting behavior, and contralateral abnormal posturing of limbs. Patients commonly have a period of postictal confusion. Rarely, this progresses to secondary generalization.[36]

Frontal Lobe Epilepsy (FLE)

Frontal lobe epilepsy (FLE), after temporal lobe epilepsy, is the second most common type of the localization-related epilepsy of childhood.[37] 

1. Ictal EEG: Scalp ictal EEG changes are difficult to appreciate with most of the frontal lobe seizures due to the movement artifacts obscuring the visibility of the underlying ictal waveforms.[38]
2. Interictal EEG: IEDs are observed in only 60% to 80% of FLE patients and have a lower localizing value than in TLE since they can be bilateral, involving multiple lobes, or even secondarily generalized.[39] The medial frontal epilepsies rarely reveal any IED, and even if they do, they are bifrontal spike and wave discharges.[40]
3. Seizure semiology: In view of the fact that the ictal EEG in FLE is of less value due to frequent muscle and motion artifacts, analyzing the ictal semiology and clinical history are important to differentiate FLE from psychogenic non-epileptic spells (PNES). This is condition is frequently misdiagnosed as epilepsy.[41]
   • Dorsolateral frontal epilepsy: The clinical manifestations of frontal lobe seizures are diverse. Seizures originating from the premotor cortex exhibit forced contralateral head deviation or head-turning called versive seizures. Activation of contralateral frontal eye field causes lateral deviation of the eye. Aphasic seizures are caused due to Broca area involvement.[42] Seizures involving the prefrontal cortex have more variable clinical manifestations. These are described as "hyper-motor seizures." These seizures usually have somatosensory auras followed by bizarre gestures, laughing, shouting, bicycle peddling, and thrashing of the extremities.[43]
   • Mesial frontal epilepsy: Associated features include loss of consciousness followed by conjugate eye and head deviation, behavioral arrest and immediate recovery of consciousness.[44]

Parietal Lobe Epilepsy (PLE)

1. Ictal EEG: Ictal scalp EEG is rarely localizing in PLE. Even extensive invasive studies done for localization can be inconclusive.[45]
2. Interictal EEG: It is common to encounter multifocal and multiregional IEDs with cortical involvement, making it difficult to localize the focus. Ristic et al. rightfully called it, 'a great imitator among focal epilepsies.'[46][47]
3. Seizure semiology: Most patients with parietal lobe seizures have no symptoms or signs suggesting parietal lobe involvement. If specific symptoms are present, they include unilateral paresthesias and pain that occur early during the seizure. Other symptoms occur due to diffuse cortical involvement and include hallucinations and distortions of space, among other symptoms.[48]

Occipital Lobe Epilepsy (OLE)

1. Ictal EEG: During the ictal period, an EEG recording is more likely to show diffuse bi-occipital activity spreading to the temporal regions than well-localized unifocal discharges in the occipital region. 
2. Interictal EEG: IEDs can occur either spontaneously or following photic stimulation. In spontaneously occurring IEDs, the discharges are in the form of unilateral posterior EEG slowing rather than spike-waves. It is also rarely localized over the occipital cortex. Instead, the posterior temporal region is the most common site. In contrast, photosensitive OLE requires intermittent photic stimulation to elicit IEDs. They occur in the form of either spikes/polyspikes confined to the occipital region or generalized spikes/polyspikes diffusely spread out to the posterior cortex.[49][50]
3. Seizure semiology: Seizure in OLE often presents with visual symptoms including hallucinations, blindness, nystagmus, and rapid blinking of the eye. Rarely, it presents as a generalized tonic-clonic seizure and impaired consciousness. This signifies a spread to neighboring cortical regions and thus makes the focus localization rather difficult.[51]

For most focal onset aware seizures, the scalp EEG commonly shows no change in simple partial seizures, because the focal ictal discharge is distant or deep, or involves too small a neuronal aggregate for a synchronized activity to register on the scalp. It becomes more evident after a larger cortical area gets involved prior to secondary generalization.

An epileptic syndrome is a chronic disorder that heavily depends on an interprofessional team to provide a holistic and integrated approach to provide the best possible long term seizure control.[52] The neurologist, epileptologist, the EEG technician, the nurse practitioner, the pharmacist, the primary healthcare provider, the neuro-radiologist and the neurosurgeon, all are an invaluable part of the interprofessional team taking care of patients with epilepsy. While the EEG technician, is key to accurate electrode placement and consequently recording an EEG without artifacts or without waveform cancellations, the neurologist and epileptologist are responsible for clinically correlating these recordings to determine the diagnosis and formulate the next course of action. The neurosurgeons play an important role in placing the electrodes for invasive monitoring for focal localization in complex cases and maximally resecting a pathological lesion associated with epilepsy while sparing the eloquent regions. Patients with epilepsy require education about their medication usage, side effects, and the importance of compliance. The clinical team comprising of the nurse practitioners, primary care physicians, and the neurologists/epileptologist play an important role in monitoring and education. The pharmacist is in charge of dispensing the correct dosage of the drug and preventing any possible drug interactions.

Collaborating by shared decision making and communication are key elements for a good outcome. The interprofessional care provided to the patient must use an integrated care pathway combined with an evidence-based approach to planning and evaluation of all joint activities.[53] The earlier signs and symptoms of a complication are identified, the better is the prognosis and outcome. [Level 3]