Abstract SNACC-86

Return to Poster Listing

Evaluation of feasibility and reliability of intraoperative visual evoked potential monitoring during craniotomy

Akasaki Y, Hayashi H, Kawaguchi M

Introduction: Postoperative visual dysfunction (POVD) is a devastating complication. Intraoperative visual evoked potential (VEP) has been considered to monitor the functional integrity of visual function during neurosurgical procedures, in which the optic pathway is at a risk of injury. Recent advances in techniques including a new light-stimulating device consisting of high-luminosity light-emitting diodes (LEDs) and induction of electroretinography (ERG) to ascertain the arrival of the stimulus at the retina have provided better conditions for stable VEP recording under general anesthesia. On the other hand, propofol anesthesia is important for successful VEP recordings because inhaled anesthetics have suppressive effect on VEP. VEP monitoring may allow us to intraoperatively detect reversible damage to the visual pathway. In this study, we investigated feasibility and reliability of intraoperative VEP monitoring during craniotomy.
Materials and Methods: Ninety patients undergoing elective craniotomy under general anesthesia from May 2009 to December 2013 were enrolled in this study. Eighty-four patients had brain tumors, two patients had aneurysms and four patients had others. Anesthesia was maintained by total intravenous anesthesia with propofol (2-3 μg/ml using target controlled infusion), remifentanil (0.05-0.3 μg/kg/min) and an additional injection of fentanyl. Intraoperative VEP monitoring was performed in bilateral eyes of all patients. ERG was recorded simultaneously with VEP to guarantee delivery of adequate flash stimuli to the retina.
For recording VEP, a luminosity of 10000-20000 Lx was applied on the eyelid at a rate of 1 Hz for 10ms and the responses were elicited from Lo, Mo, Ro (international 10/20). The needle electrodes were inserted subcutaneously. Averaging runs of 100-150 stimuli were adequate to generate clear VEP responses. VEP amplitudes were defined as the range of peak-to peak amplitude between N75 and P100. More than 50% decrease in VEP amplitudes compared to the control level was defined as significant VEP change. All patients were performed with ophthalmological examination for visual field and visual acuity preoperatively and postoperatively. We evaluated the success rate of VEP recording and examined the correlation between intraoperative significant decreased VEP waveform change and postoperative visual dysfunction. Accuracy was defined as the sum of the true positive and true negative cases out of all cases.
Results: Fifty-four had preoperative ophthalmological complications including defects of visual field and declined visual acuity. The success rate of recording control VEP was 94.4 % (85 of 90 cases). Of the 85 patients, continuous significant decreased changes of VEP amplitude were observed in 8 patients without POVD (false positive). One patient had new postoperative visual field deficit without intraoperative reduction of VEP amplitude (false negative). As a result, the false positive, false negative and accuracy of intraoperative VEP monitoring during craniotomy ware 9.4%, 1.2% and 89%, respectively.
Conclusion: Intraoperative VEP monitoring may be feasible with satisfying reliability to predict postoperative visual dysfunction during craniotomy under general anesthesia.

Back to Top