An ideal skull base reconstruction by our cognition should implement philosophies including: 1. Separating intracranial space from aerodigestive tract. 2. Promoting rapid and stable adhesion, cicatrization, and epithelialization at defect site. 3. Restoring the original anatomical layers of bone and dura. 4. Excellence in feasibility, safety, and patients’ subjective experience [4, 6, 23, 24]. According to these points, we have launched CEFB application since 2015. It may be unusual that our study included substantial pituitary adenoma cases because adenoma of epidural origin usually has low rate of intraoperative CSF leak . However, as a senior adenoma center covering 30 million population, our institution admitted high proportion of giant or invasive adenoma. Due to the compromised or breached diaphragm sellae in these patients, the morbidity and severity of intraoperative CSF leak are higher than general level. Percentage of grade III leak featured by huge diaphragma defect and direct opening to suprasellar cisterns is 48.5% in all included cases. The baseline data distribution in groups of CEFB and PNSF had statistically confirmed consistency, which provided solid comparability for two methods.
The purposes of inlay in CEFB
Multilayer reconstruction has become a commonly accepted principle in EES [14, 25-28]. The inlay materials are not adequate for CSF blockage, but they contribute to reducing leak size and CSF flow, eliminating dead space, buffering impact of CSF pulsation, and alleviating CSF pooling or soaking [29, 30]. The key point of CEFB inlay procedure is that the volume of fat graft should be finely adjusted to obtain optimal subdural tension fitting the buttress pressure of wedged bone graft, generating appropriate tightness of attaching between fascia and dura. Either excessive or inadequate pressure would hinder the reconstruction stability, which is illustrated by the reoperation of one failure case in CEFB group described above.
The benefits of partial dural suturing in CEFB
With the advances of surgical technology, the deep suturing and knotting are no longer big issues in EES, but the literally “watertight” suturing is still technically challenging due to the dural dehydration, fragility, or electrocauterization [8, 31]. The partial suturing was not enough to seal CSF leak, but we deem it still has benefits as: 1. Reducing the dural defect and confining it under the center of rigid buttress. 2. Providing dural interface for onlay fascia to attach, avoiding direct contact between fascia and inlay grafts. 3. Anchoring ADM and fat in place and avoiding them fall into suprasellar cistern or ventricle . 4. Increasing intrasellar tension and compactness .
Comparison between CEFB and gasket-seal
CEFB shares the same concept of rigid reconstruction with other methods represented by gasket-seal  and ISBF . CEFB has the similarity of “hard material buttressing soft structure” with those two techniques, but with significant modifications. As its name addressed, gasket-seal focuses on the watertight closure of defect. The circumferentially wedged material works as a plug wrapped by fascia to increase sealability, so its shape are required to be highly matched to the shape of bone defect [6, 33]. Artificial material is frequently used [6, 19, 34], because the autologous bone graft does not always perfectly fit the defect. The core of CEFB is the buttress pressure firmly holding fascia lata on the dura to promote tight attaching and mutual adhesion. Firstly, bone flap in CEFB was embedded at only two sides of bone defect, instead of fully circumferential wedging. The size of bone flap is qualified by adequate length on only one axis. This is an easier requirement for bone graft harvest, which promotes the usage of autologous bone in more cases, and brings lower costs and risks of infection or rejection. In our study, 67.2% (133/198) of all cases have acquired satisfactory autologous bone harvest. Secondly, the two-side embedding of CEFB allows fascia lata to stretch out through the gaps on non-wedging sides and to be paved smoothly on skull base. But in gasket-seal, the fascia lata formed “cauliflower leaf” shape [6, 33] due to the tight depression in the center . The tilted or curled edge of fascia is hard to be flattened to provide smooth attaching. Thirdly, the partial dural suturing ensures the correct epidural embedding of bone flap, avoiding it being accidentally misplaced into subdural space. The firm contacting between sutured dura and fascia also assists the adhesion formation. Finally, the gasket-seal is not ideal in case of defect traverses two geometric planes, because the rigid buttress is not curved to fit the contour of defect . In our EEEA cases with sufficient bone graft harvest, we used two separated bone flaps to wedge at different defect planes respectively (Fig. 3a). The fascia lata then could be supported evenly and hold in place on angled planes.
Comparison between CEFB and ISBF
ISBF is another typical method of rigid buttress reconstruction. Its advantages are eminent in perfect contour match and tissue compatibility . The forming of ISBF is by microdrill of 2.5mm diameter diamond burr, which inevitably inflicts loss of bone substance. Although it can be adjusted to get several wedging points, the stability of this fixation is limited. Besides, ISBF is tend to be “placed” on the defect plane, rather than countersunk below it. The buttressing pressure applied by ISBF is weaker compared to the firm wedging in CEFB. Moreover, the fundamental requirement of ISBF is the intact of skull base bone structure without tumor invasion. This indication may be easy to meet in cases of suprasellar tumor, but high proportion of pituitary adenoma in our case series often involved infiltration or sabotage of bone.
Resistance of CEFB against counteracting forces
The forces of brain gravity, CSF pulsation, and intracranial pressure are great concerns in skull base repair. These downward forces are prone to incur compromise of reconstruction . Hence, iodoform gauze packing, lumbar drainage, and intranasal balloon et al are used as countermeasures [6, 34, 35]. The integrity of CEFB is sturdy enough to resist these forces due to the rigidity of firmly wedged bone flap. No dislocation or fracture of bone flap was observed in all of our CEFB cases, even without iodoform gauze packing or lumbar drainage. These downward forces might be conjecturally helpful to form more solid compression of CEFB structure, to strengthen the watertight and attaching. This mechanism possibly contributed to the shortened bed stay time in CEFB group. Especially in cases with preoperative hydrocephalus, either CEFB alone or combination with PNSF exhibited good outcome of 0 postoperative CSF leak in 15 cases.
Considerations about PNSF application
PNSF is a milestone in development of skull base reconstruction for its fast healing and long-term security of closure . But the harvest of PNSF necessitates long incision on nasal septum to mobilize extensive piece of mucosa, and migrate it to skull base. The exposed donor site needs 6-12 weeks to be re-epithelialized . Nasal complications related to this extensive shift of mucosa are not rare [15-17, 36]. The nasal packing of iodoform gauze or balloon regularly used in associate with PNSF also influences mucosa regeneration and patients' subjective experience due to the intranasal pressure and stimulation . Garcia-Navarro et al reported PNSF did not make statistical difference in their gasket-seal practition, which raised the question on PNSF necessity . In our study, it also showed no significant difference in postoperative CSF leak and infection between CEFB and PNSF, and nasal complications are significantly reduced in CEFB group. The effectiveness of CEFB is comparable and proximate with other representative reports on postoperative CSF leak (Table 4). Our data suggests that in presence of appropriate CEFB with multilayer reconstruction, PNSF may not be a sole mandatory option. Pros-cons of PNSF should be individually weighed, and any form of PNSF “abuse” is not justified. We do not attempt to simply replace PNSF with CEFB. On the contrary, CEFB is not exclusive but highly compatible with PNSF. In 23 cases of grade III CSF leak with multiple risk factors, the combination of 2 techniques achieved 0 postoperative CSF leak. It implied that with the additional protection from PNSF, CEFB could provide reliable security in high risk cases.
Limitations of CEFB and present study
First, harvest of bone flap is not always sufficient or available due to anatomical variations or tumor invasion, especially in case of oversize defect extending laterally. Second, the coverage of fascia under the wedging edges of defect is limited. Several measures could be taken to improve the limitations: 1. During surgical approach, any bone structures with utilizing potential should be preserved by avoiding excessive grinding of microdrill. 2. Bone graft could be tailored into narrow strips to intervally wedge at defect (Fig.3b). 3. The wedging sides of defect could be carefully enlarged to expose more dural surface for fascia attaching (on premise of guaranteeing embedding stability). 4. Dural defect could be minimized by improving dural incision design and suturing. Finally, in the worst scenario, when all measures turn out to be unassuring, PNSF remains the trustworthy resort.
The number of cases in our study is still limited, and the establishment of randomized control is difficult. Current follow-up period of 6 months is also not enough to perform long-term evaluation. More rigorous study design and accumulation of cases are required for further assess of CEFB technique.