Ventriculoperitoneal shunt failure is most often caused by catheter obstruction, and freehand insertion at Keen's point leaves the tip free in cerebrospinal fluid in only about one case in ten. To improve this, the authors built a patient-specific digital twin from preoperative CT, reconstructing the skull, scalp, ear, and ventricles with deterministic geometry. The system searches every possible entry around Keen's point for the perpendicular trajectory with the longest intraventricular length under safety constraints, then returns the plan as two ruler measurements from the ear and two depths.
In an IDEAL Stage 2a feasibility study, 11 consecutive adults underwent freehand insertion using these plans. Ten of 11 achieved Hayhurst Grade 1 placement, compared with 22 of 213 historically (p = 9 x 10^-9). Median entry deviation was 2.5 mm and angular deviation 5.9 degrees, and the catheter had 1.4 cm more intraventricular length than with Keen's point. No intraoperative complications were reported.
The surgical technique itself was unchanged; the improvement came from choosing entry point and depth per patient. The authors state that whether this approach lowers revision rates is the hypothesis for the next stage. As a small, early-stage study, the results are promising but require validation in larger trials.