Deep-seated brain tumors are often best treated by primary surgical excision. Traditional microsurgical techniques can cause retraction injury and require extensive brain dissection. To mitigate this risk, stereotactic-guided tubular retractors were developed; however, the risk of shear injury remains. We created a stereotactic-guided dilatable port system to create a corridor for deep brain tumor surgery along the trajectory of a brain needle to minimize iatrogenic brain injury. Of the 8 included patients (6 colloid cysts, 1 metastasis, 1 intraventricular meningioma), 5 had undergone frameless and 3 frame-based stereotactic targeting. We used a tans-sulcal trajectory and a 2.6-mm stereotactic needle. At the target depth, the cannula was removed and the balloon inflated to 14 mm. The balloon was deflated and removed before placing the port. Pre- and 3-month postoperative magnetic resonance imaging scans were used to measure the T2-weighted signal change and residual cannulation defect. These patients were compared with a case-matched standard endoscopic port surgery cohort. All patients had undergone total lesional resection without new neurologic deficits. Patients undergoing dilatable endoscopic port surgery (DEPS) had significantly smaller residual cannulation defects (P < 0.05) but no significant differences in postoperative T2-weighted signal changes or diffusion restriction volumes at 3 months postoperatively (P > 0.05). DEPS might be a safe alternative to standard endoscopic port surgery or microsurgery for deep-seated brain tumors. The degree of iatrogenic injury using DEPS, as determined by magnetic resonance imaging analysis, might be equivalent to or less than that with standard port surgery techniques, although larger sample sizes are needed for validation.