Bispectral index as a possible early marker of cerebral hypoperfusion.

We report a case of unexplained transient ventricular fibrillation in a child with ventricular septal defect (VSD) scheduled for VSD closure. The bispectral index (BIS) values dropped within 15s of transient cardiac arrest and it showed recovery within 15 s of restoration of circulation. Monitoring BIS during surgery, especially cardiac surgeries may help identify periods of cerebral ischemia early.

CASE HISTORY

A 7-year-old male child weighing 20 kg with a perimembranous ventricular septal defect (VSD) was posted for VSD closure under cardiopulmonary bypass.The preop ECG showed normal sinus rhythm with evidence of left ventricular hypertrophy (LVH) and echocardiography showed a 5.5 mm perimembranous VSD with mild enlargement of left atrium (LA) and left ventricle (LV). Other preop investigations including serum electrolytes and renal function tests (RFTs) were within normal limits.

The child was pre-medicated with tab. ranitidine 75 mg on the night before as well as on the morning of the surgery. Five lead ECG and pulse oximetry were placed and the room air saturation was found to be 100%. The child's hemodynamic and other parameters were being routinely recorded into an online database which could record the bispectral index (BIS) values at 5 s intervals and the hemodynamic variables at every 10 s interval. The child was then induced through a pre-existing functional 22G cannula with inj. fentanyl 100 μg, inj. thiopentone 80 mg, and inj. vecuronium 2 mg following which inhalational halothane was started. A 22G arterial cannula was placed in the left upper limb and transduced and the opening arterial pressure was 127/98 mmHg.

The airway was secured with a 6mm cuffed endotracheal tube (Portex™) and fixed at 17 cm. There was no difficulty in intubation and on auscultation air entry was found to be equal. Under aseptic precautions, central line was placed in the right internal jugular vein (IJV) and sutured in place at 8 cm at skin level. Subsequently, as child was being positioned for surgery he had a sudden episode of hypertension (148/70 mmHg) and ventricular tachycardia/fibrillation which rapidly degenerated into asystole. The problem was promptly recognized. Chest compressions were started and the child was taken on bag ventilation and the defibrillator was charged and readied. The rhythm reverted back to sinus with compressions and the entire episode lasted for 40 s. During the episode, halothane was switched off and subsequently isoflurane was started post recovery from the episode.The immediate post event arterial blood gas (ABG) was within normal limits (7.37/464.7/42.5/138/3.4/25/39%). The child subsequently had a stable intra-op and post-op course and was discharged.

The online monitoring data was reviewed and 5 s BIS and 10 s hemodynamic and other parameters were reviewed. The BIS values during the event dropped from a value of 63 to a value of 39 within 15 s of the onset of asystole and similarly in the recovery phase picked up from a value of 35 to 65 in 15 s. This finding suggests that the BIS values could be an early possible marker of cerebral hypoperfusion. Figure 1 depicts the changes in the hemodynamics and BIS during the period in question.

Figure 1

Changes in hemodynamics and BIS during the said period. Note the drop in BIS followed by recovery. Closely monitoring BIS might detect cerebral hypoperfusion early

DISCUSSION

Though we were not able to arrive at the exact cause for the event, we suspect that it could be due to an accidental drug bolus or a flush of the inotrope line. The epinephrine line was connected to the central venous pressure (CVP) port and accidental flush through this line may have lead to the event. This is a strong possibility as the child was also receiving halothane anesthesia and this might have precipitated the event.

In our institution we routinely record the monitoring data onto an online database which records the BIS values at 5 s intervals and hemodynamic variables at 10 s intervals. Review of this record facilitated a greater understanding of the sequence of events and the trends over this period.

BIS is a processed electroencephalography (EEG) parameter with clinical validation and has traditionally been used for monitoring the depth of anesthesia during surgery. The BIS index is the output from advanced EEG analysis. It is derived utilizing a composite of measures from EEG signal processing techniques such as spectral analysis, power analysis, and time domain analysis.[1] These measures were combined with an algorithm to reflect the clinical state of the patient. The index is a dimensionless number scaled to specific clinical endpoints and specific EEG features. Awake, unsedated patients typically have BIS values of 97 and above.[1] BIS values of 60 correlate with general anesthesia and values less than 30 imply significant amounts of EEG suppression and a value of 0 indicates an isoelectric EEG.[1]

BIS systems display the index as a single value, calculated from data collected over the last 15-20 s of EEG recording and updated every second. Determining the value over several seconds helps smooth the data so as to prevent excessive fluctuations in values. The BIS value while extremely sensitive may still not change instantaneously with changes in clinical situation such as during anesthetic induction and emergence and lags by 5-10 s.[1]

Sharp decreases in BIS value are known to be associated with hypoglycemia, hypothermia, cerebral ischemia, hypoperfusion, and cardiac arrest.[1] This has been attributed to decrease in cerebral metabolism which leads to changes in EEG and hence in BIS values. Patients with neurological dysfunction may have low BIS values and anticonvulsant drugs may reduce BIS values.[1] A number of articles have already described the use of BIS as a possible marker of cerebral hypoperfusion. The exact time frame in which these BIS changes may correlate with periods of ischemia has not been studied or documented.

Myers et al.,[2] found that a time compressed Fourier analysis of EEG could be used to monitor cerebral perfusion during carotid endarterectomies (CEAs) and during open heart surgery. Azim et al.,[3] reported a drop in the BIS values during a post cardiac surgery cardiac arrest and the improvement in the BIS values with cardiopulmonary resuscitation (CPR). Morimoto et al.,[4] reported drops in BIS value during periods of decreased cerebral perfusion pressure in a patient undergoing surgery for arteriovenous (AV) shunt.

Trojaborg et al.,[5] recorded the EEG in 52 patients with cerebrovascular disorders slated for reconstructive surgery. Changes in cortical activity during a 2 min test occlusion of the carotids were studied. Slowing of the EEG was seen when the blood flow decreased to 16-22 ml/100g/min and mean pressure in the internal carotid dropped to between 29-50 mmHg. Flattening of the EEG was seen when blood flow decreased to 11-19 ml/100g/min and when mean pressure dropped to 15-46 mmHg. Hayashida et al.,[6] similarly found decreases in BIS values and cerebral hemoglobin saturation during periods of hypotension during pediatric cardiac surgery.Hemmerling et al.,[7] found BIS to be an indicator of cerebral hypoperfusion during off pump coronary artery bypass graft (CABG). Bleeker and colleagues.,[8] also found that BIS values dropped in a patient who suffered from a transient ischemic attack (TIA) prior to induction for CABG.

Some studies however, have not found significant correlation between BIS values and cerebral hypoperfusion. Deogaonkar et al.,[9] studied the reliability of BIS as a marker of intraoperative cerebral ischemia during awake CEA and found that BIS did not correlate well with cerebral ischemia. In this study, 52 patients undergoing awake CEA were monitored with BIS along with neurological assessment (upper and lower limb strength and the verbal component of the Glasgow coma scale for speech). Overall mean BIS values were 96. In five patients who showed clinical evidence of cortical ischemia during cross clamping, there was no change in the BIS values throughout the procedure (96.7). In one patient the BIS value decreased to 38,5 min after incision, but recovered within the next 10 min. The study could not find a reliable correlation between BIS and periods of cerebral ischemia.

Despite a number of studies documenting the EEG and BIS changes with cerebral hypoperfusion, no study has so far commented on the time frame within which the BIS value drops. Awareness of this time lag may help in early detection of the period of cerebral hypoperfusion/ischemia and may help in early institution of measures to correct the cause of the ischemia. Determination of the time lag by experimentation is not feasible due to ethical issues involved. In the present case, the BIS value dropped over a 15 s interval and was restored within 15s of the restoration of flow and provided an indirect evidence of the time lag from onset of asystole to changes in BIS values. Though ECG and direct arterial pressure monitoring may provide earliest indication of dangerous hemodynamic disturbances, the early drop in BIS may also help in early institution of measures to rectify the causes leading on to cerebral ischemia and may also serve as an indicator of the adequacy of cerebral flow.

BIS value is essentially derived from the frontal leads and ischemia/hypoperfusion of the frontal cortex will get reflected directly in the BIS values. Apart from this hypoperfusion of brainstem may result in a decrease in BIS values due to multiple afferent projections from brain stem nuclei to the cortex.[8] Abnormal EEG as in patients with dementia, cerebral palsy, postictal state, and severe brain injury may all result in low BIS values.

Though BIS values may be influenced by a number of patient related factors, anesthetic agent related factors, and may also be influenced by medical devices; however in the light of the available evidence, BIS seems to be a fairly reliable and early marker of cerebral hypoperfusion.