Free Fat Grafts in Endoscopic Skull Base Closure: A Clinical Outcome Analysis.

Background

Autologous fat grafts are widely used for skull base reconstruction following endoscopic endonasal surgery. However, uncertainty regarding their postoperative volumetric evolution may complicate the distinction between expected postoperative changes and residual or recurrent disease on imaging.

Methods

We performed a retrospective volumetric imaging analysis of patients undergoing endoscopic endonasal skull base surgery with reconstruction using autologous fat and the 3F technique between 2013 and 2023. Fat graft volumes were measured on postoperative day 1 CT scans and on 3-, 9-, and 15-month postoperative MRI using standardized volumetric segmentation analysis (method described in detail in the main text). It should be noted that baseline measurements were derived from CT, whereas follow-up assessments were performed using MRI. Resorption rates were correlated with demographic, pathological, surgical, and postoperative variables.

Results

Thirty-four patients met inclusion criteria. Mean initial fat graft volume was 3.01 ± 2.65 cm3. Overall, fat graft volume demonstrated a consistent temporal decline, with mean reductions of 56.8% at 3 months, 75.3% at 9 months, and 81.8% at 15 months. In subgroup analysis, differences in resorption were observed according to surgical approach, with higher resorption following transsellar compared with transtuberculum approaches (87.4% vs. 74.8% at 15 months, p = 0.042). These findings were closely related to graft compartment, although compartment and surgical approach showed substantial overlap in this cohort. No significant associations were detected between resorption rate and age, sex, comorbidities, postoperative CSF leak, extent of resection, or adjuvant radiotherapy.

Conclusions

Autologous fat grafts used for skull base reconstruction demonstrate substantial early postoperative resorption followed by slower progressive volume loss. Recipient anatomical compartment was associated with differences in observed resorption patterns, although this relationship should be interpreted in the context of overlap with surgical approach and limited sample size. These findings may assist in improving interpretation of postoperative imaging by clarifying the expected temporal pattern of fat graft evolution after endoscopic skull base reconstruction.