LAUSR

OBJECTIVE Catheter malposition represents one of the major causes of ventriculoperitoneal (VP) shunt dysfunction. The usefulness of intraoperative fluoroscopy using skull landmarks has already been proved to decrease catheter malposition and surgical revision rates. After introducing intraoperative computed tomography (iCT) in our department, our objective was to evaluate the accuracy of this imaging modality to decrease cranial catheter misplacement compared with intraoperative fluoroscopy. METHODS In our retrospective analysis of 152 patients, catheter placement was evaluated by iCT (n = 48) and biplane fluoroscopy (n = 57). A control group (n = 47) had no intraoperative imaging. Outcome measures included accuracy of ventricular catheter position, revision surgeries, and clinical outcomes. RESULTS Ventricular catheter placement was accurate in 24/48 patients with iCT and 45/57 patients with fluoroscopy (P = 0.002) versus 23/47 patients in the control group. Sensitivity and positive predictive value for estimating optimal catheter position with iCT were 100% and 54%. The specificity and negative predictive value were 50% and 100%. After intraoperative revision, 4 catheters remained malpositioned in the iCT group, whereas the fluoroscopy group had none (P = 0.03); 2 of these 4 catheters were revised postoperatively. CONCLUSIONS Fluoroscopy may be the method of choice to intraoperatively assess ventricular catheter positioning. In our experience, iCT shows a tendency to be more time consuming and, in the beginning, was not associated with a steeper learning curve. Another consideration was the significant higher radiation exposure per patient. iCT did not improve the accuracy of catheter placement and did not decrease early revisions for VP placement patients.