Using a New EEG Based Algorithm for Intra-uterine Fetal Brain Monitoring.

Fetal heart rate (FHR) monitoring is commonly used in order to reduce fetal intrapartum asphyxia or any associated long-term disabilities. The electronic fetal monitoring (EFM) was first described by Hon and others during the 1960s (1) (2) and introduced into clinical practice before any clear evidence of benefits had been demonstrated.

The EFM records the fetal heart rate, and characteristics such as baseline, accelerations, decelerations and variability are constantly evaluated in order to assess the hypoxic condition of the fetus and to identify fetal distress.

Nevertheless, most of the studies (3-5) demonstrated that abnormal patterns of the fetal heart rate are of low predictive value for intrapartum fetal hypoxia or metabolic acidosis. Among high-risk pregnant woman, the probability of fetal metabolic acidosis in the presence of repeated, variable decelerations is 25%, and 48% in the presence of late decelerations (6). According to the Cochrane review (7) EFM didn't decrease the rates of cerebral palsy (CP), asphyxias complications or perinatal morbidity. The positive predictive value of non-reassuring FHR patterns for the prediction of CP among singleton newborns with birth weight of 2500 grams or more is only 0.14% (8).

Hypoperfusion to a certain brain areas involves the rapid loss of brain function due to disturbances in brain blood supply. EEG shows typical changes in brain ischemia. These changes include: attenuation of faster waves, particularly in the beta and alpha frequency bands; enhancement of slower waves, mainly in the delta frequency band; regional attenuation without delta enhancement (RAWOD) and reduction of variability in the EEG signal (9). However, reaching effective sensitivity and specificity in immediate brain ischemia detection on the basis of EEG changes is still considered a challenge, as it depends on the presence of a skilled expert in neurophysiology to analyses raw EEG data in real-time.

The investigators developed a unique platform algorithm, which analyses EEG data, and recognizes patients suffering from brain ischemia (stroke), based on the phenomena of interhemispheric decreased synchronization. Interhemispheric synchronization is the measure of correlation in EEG activity between ipsilateral and contralateral hemispheres, where values range between 1 (complete synchronization) and 0 (no synchronization). Increased synchronization means greater correlation between related ipsilateral and contralateral activities and vice versa for decreased synchronization. Normal brains should have a high correlation/ synchronization. Once brain ischemia occurs and the hemisphere is damaged, the interhemispheric synchronization decreases. In a recent study, we have shown the decrease in synchronization when a certain part of the brain is anaesthetized, causing a lesion similar to brain ischemia, and the recovery of synchronization, when the anesthesia recovered2. In another study, the investigators have tested the reliability of the algorithm in patients undergoing mechanical thrombectomy for acute stroke. In the study patients with acute stroke admitted to neuro angio intervention undergone short EEG test for 5 minutes. At the end of the procedure patients underwent a second short EEG test. In the test the interhemispheric synchronization were analyzed. It was shown that patients admitted to the neuro-angio intervention, who had brain tissue that might be salvable, had an average synchronization of 0.8. If the procedure succeeded and there was a clinical improvement - the synchronization remained high. If the procedure did not succeed, or succeed without clinical improvement, the synchronization decreased to an average of 0.65. Importantly, patients who were not candidates for neuro intervention, because they didn't have salvable brain, their index of synchronization was low to begin with (figure 1a and 1b). This preliminary assay proves the ability of the synchronization index to detect brain hypoperfusion that occurs, for example, in stroke.