Skull base defects: outcomes and reconstruction from a tertiary center in India
Article information
Abstract
Background
Skull base defects, accidental or planned during tumor resections, present significant reconstructive challenges. Effective repair prevents cerebrospinal fluid (CSF) leaks, meningitis, and pneumocephalus.
Methods
This retrospective observational study analyzed 26 patients with anterior or lateral skull base defects treated between 2018 and 2024 at a tertiary care center in Western India. Both planned iatrogenic and accidental intraoperative defects were included. Patient demographics, surgical approach, defect characteristics, reconstruction techniques, and postoperative outcomes were recorded. All patients underwent multilayered repair with tailored perioperative management, including selective lumbar drainage.
Results
Among 26 patients, 14 (53.8%) had anterior and 12 (46.2%) had lateral skull base defects. Intraoperative CSF leak occurred in 11 patients (42.3%). Defects were repaired using combinations of autologous tissue (fat, fascia lata, muscle), allogenic grafts (DuraGen), and vascularized flaps. Seven patients (26.9%) required lumbar drains. All reconstructions were successful with no secondary failures. However, two major complications were observed: one case of pneumocephalus and one of postoperative vision loss following ossifying fibroma excision. No postoperative meningitis or recurrent CSF leaks were noted.
Conclusion
Meticulous multilayered repair strategies and structured perioperative protocols provide excellent outcomes in skull base defect reconstruction. Differentiating between planned and accidental iatrogenic defects allows for better preoperative planning and individualized management. Despite a 100% repair success rate, rare but serious complications such as vision loss highlight the need for careful surgical technique near critical neurovascular structures.
INTRODUCTION
Skull base defects may arise from traumatic and atraumatic causes, including iatrogenic injuries during surgery. Pathologies requiring extensive anterior or lateral skull base resections demand complex reconstruction tailored to defect size, location, disease type, and surgical expertise [1]. Historically, open approaches were the mainstay for tumor extirpation at the skull base, though associated with significant morbidity, including cerebrospinal fluid (CSF) otorrhea/rhinorrhea, pneumocephalus, and meningitis. The importance of restoring the barrier between surgical cavities and intracranial structures led to the evolution of skull base reconstruction techniques. Endoscopic approaches, now preferred over open techniques, have further advanced this field. A watertight closure is critical, particularly to reduce the risk of ascending infections and avoid delays in adjuvant therapies for malignant disease. Effective repair also shortens hospital stay and reduces reoperations [2].
Initial endonasal endoscopic approaches were associated with postoperative CSF leak rates up to 30%–40% [3,4], which decreased significantly to approximately 5% after introducing the vascularized nasoseptal flap [5,6]. Despite these advances, CSF leak remains a common and costly complication [2,7]. However, limited real-world data exist on reconstruction techniques and outcomes in diverse skull base defects.
This study presents a retrospective case series involving anterior and lateral skull base defects managed using various reconstructive approaches, including endoscopic, microscopic, and combined techniques. It aims to evaluate surgical outcomes and provide insight into the practical application of multilayered repair strategies in a tertiary care setting.
METHODS
This retrospective observational study was conducted in the ENT (otorhinolaryngology) department of a tertiary care hospital in Western India from 2018 to 2024, following approval from the Institutional Ethics Committee (IEC No. AIIMS/IEC/ 2024/5286). It adheres to the STROBE guidelines. All cases involving microscopic or endonasal endoscopic or combined approach lateral and anterior skull base surgeries were reviewed.
Inclusion criteria
Patients aged ≥10 years with anterior or lateral skull base defects caused by planned iatrogenic interventions (e.g., tumor excision, mucormycosis debridement) or accidental iatrogenic injuries (e.g., CSF leak during functional endoscopic sinus surgery or mastoidectomy) were included. Surgeries performed via endoscopic, microscopic, or combined approaches were considered. Defects were classified intraoperatively as small (<0.5 cm), intermediate (0.6–2 cm), or large (>2 cm).
Exclusion criteria
Cases with congenital defects (e.g., meningoceles), traumatic skull base fractures without surgical intervention, spontaneous CSF leaks, or incomplete records were excluded.
Data collection
The demographic details, skull base defect site and size, repair method and subsequent postoperative complications in the follow-up period were noted. In all the cases, the skull base defect was initially identified by inspection for dural pulsations and then intraoperatively with the Valsalva manoeuvre by the anesthesiologist [2] to look for any CSF micro leaks. We divided the defects based on size, as observed intraoperatively by the operating surgeon, considering defects up to 0.5 cm as small, 0.6–2 cm as intermediate size and more than 2 cm as large defects. The various modalities used for measuring the size intraoperatively were the instruments routinely used in skull base surgeries–suction tip (size 5), 15 mm and 30 mm angled ball probes.
Reconstruction techniques
The modalities used for repair included free mucosal flaps, Surgicel (Absorbable hemostat, oxidized regenerated cellulose), Gelfoam (absorbable gelatin sponge), Tisseel glue (fibrin sealant from Baxter), tensor fascia lata/fat/muscle, DuraGen Plus (adhesion barrier matrix), depending on the size and complexity of the defect. Larger or complex defects required nasoseptal flaps, temporalis muscle rotation, osteoplastic flaps, or cul-de-sac closures for lateral defects.
Postoperative management
The patients were observed postoperatively for altered sensorium, vomiting, and signs of meningitis, such as neck rigidity/fever/Kernigs or Brudzinski signs. Routine postoperative ophthalmological examination was done to check visual acuity and rule out papilledema. Any of the complications mentioned above was an indication for further intervention, either medical or surgical. Immediate postoperative radiology (computed tomography [CT] or magnetic resonance imaging [MRI]) was done in cases where a large surgical defect was repaired, or where the pathology was preoperatively extending intracranially, to rule out increased intradural pressure or collection and in cases where residual pathology was suspected. All precautions needed for the repair to stabilize and prevent any failure and subsequent CSF leaks, such as head end elevation, stool softeners, limited ambulation, and acetazolamide administration, were followed postoperatively. Lumbar drain was inserted with intermittent CSF drainage for 72 hours, only if intraoperatively a high-pressure leak was observed, large or multiple skull base defects were present, or if there was a doubt of CSF microleaks post-reconstruction.
RESULTS
A total of 26 patients were included in the study, of which 14 patients (53.8%) had anterior skull base defects. These defects were a direct result of surgical intervention to remove a pathology. Of 26 patients, 13 (50.0%) were men, and the rest were women. The age of the patients ranged between 10 and 74 years. Patients included had a variety of pathologies, from benign entities such as ossifying fibroma and juvenile nasopharyngeal angiofibroma, to malignancies with skull base involvement. Few were post-surgical candidates with skull base defects with or without brain herniation or CSF leak (Tables 1, 2).
Clinical and surgical characteristics of patients with skull base defects
Summary of patient demographics, defect characteristics, and reconstruction techniques
Intraoperative CSF leak was identified in 11 of 26 patients (42.3%). The skull base defects were classified based on defect size into small-sized (n=4), intermediate-sized (n=16) and large-sized defects (n=6). The standard method to repair most defects was a multilayered reconstruction using fat, tensor fascia lata and Gelfoam-Surgicel pledgets supported by Tisseel glue. Wherein the defect was larger and more complex, DuraGen, Naso septal flap reconstruction or temporalis muscle rotation, was incorporated. Lateral skull base defects had an additional osteoplastic flap closure (surgical technique where a section of bone is temporarily removed and then repositioned to reconstruct a defect) in such cases (Figs. 1–5). Cul-de-sac closure was incorporated for lateral skull base malignancies when there were multiple defects or a risk of CSF leak in a large postoperative cavity. Seven patients (26.9%) required postoperative lumbar drain insertion. All 26 reconstructions were successful with no secondary failures observed during follow-up; however, the limited sample size restricts the generalizability of this finding.
A 41-year-old man with esthesioneuroblastoma Kadish C who underwent endoscopic excision (Case 3 in Table 1). (A) Computed tomography (coronal section) showing the extent of an esthesioneuroblastoma involving the ethmoid and sphenoid sinuses with skull base erosion. (B) Endoscopic view showing the skull base defect after tumor excision prior to flap coverage. (C) Nasoseptal flap reconstruction of the skull base defect providing vascularized coverage and watertight closure.
A 25-year-old woman with esthesioneuroblastoma Kadish D, who underwent combined approach excision (Case 4 in Table 1). (A)Combined cranio-endonasal approach used for excision of esthesioneuroblastoma, providing exposure to the anterior skull base. (B) Application of DuraGen matrix over the dural surface following tumor resection to reinforce the skull base repair. (C) Tensor fascia lata graft placement over the skull base defect for multilayer reconstruction after tumor excision.
A case of 47-year-old woman, who underwent endoscopic excision of ossifying fibroma (Case 8 in Table 1). (A) Axial, sagittal, and coronal computed tomography (CT) images show bifrontal air pockets post excision of ossifying fibroma. (B) CT scan showing the maximum width of the pneumocephalus at the right sided frontotemporal region.
A 30-year-old woman, post canal wall down mastoidectomy, with meningoencephalocoele (Case 15 in Table 1). (A) High-resolution computed tomography (coronal view) showing a lateral skull base defect with post-surgical meningoencephalocele following canal wall down mastoidectomy (yellow arrow). (B) Intraoperative view demonstrating excision of meningoencephalocele and identification of the skull base defect (yellow arrow). (C) Reconstruction of lateral skull base defect using an osteoplastic flap (black asterisk) after excision of meningoencephalocele, achieving multilayer closure. Postoperative outcome: intact closure, no cerebrospinal fluid leak on follow-up.
Complications
Importantly, two significant complications—one case of postoperative vision loss and one case of pneumocephalus—were encountered, highlighting the potential risks associated with skull base surgery. An 11-year-old boy with a large juvenile ossifying fibroma involving the sphenoid and posterior ethmoid sinuses, developed postoperative vision loss. The tumor measured approximately 7×6 cm and was compressing both optic nerves with radiological evidence of grade IV papilledema preoperatively. Endoscopic endonasal excision was performed, during which extensive drilling was required near the optic canal and skull base. Postoperatively, the patient developed complete vision loss in the right eye (perception of light negative), while the left eye retained normal vision (6/6). Fundus examination revealed bilateral grade II disc edema. Imaging demonstrated edema around the optic nerve without disruption of the bony canal. Despite prompt ophthalmological management and medical therapy (including corticosteroids and acetazolamide), vision did not recover in the affected eye.
A 35-year-old woman with sinonasal ossifying fibroma developed postoperative pneumocephalus. Initial CT on postoperative day (POD)1 revealed mild bifrontal pneumocephalus (max thickness 1.4 cm). Subsequent imaging on POD5 showed enlargement (right frontal pneumocephalus 4.5×5.1 cm), but the patient remained neurologically stable. Conservative management was instituted with oxygen therapy, head elevation, and lumbar drainage (dislodged on POD2). Serial CTs demonstrated a gradual reduction, with pneumocephalus measuring approximately 3.0×1.0 cm by POD9. She was also treated for a postoperative febrile episode with intravenous Meropenem. She was clinically stable at discharge, tolerating oral feeds, with preserved vision (finger counting at 3 m bilaterally) and no neurological deficits.
Postoperatively, the patients of both anterior and lateral skull base defects were kept on close watch during their entire tenure of admission and were strictly followed up on an outpatient department basis to assess for any late postoperative complications. Routine examination and endoscopic/microscopic assessment confirmed the integrity of the repair method and ruled out any infection or CSF leak.
DISCUSSION
CSF leak remains the most frequent complication in endoscopic endonasal skull base surgery. Before the introduction of vascularized nasoseptal flaps, postoperative CSF leak rates ranged from 15% to 20% [3]. Although intraoperative leaks were identified in 42.3% of our cases, none developed postoperative leaks—likely due to meticulous multilayered repair and strict perioperative protocols including immobilization, diuretics, and antibiotics.
The risk of ascending meningitis post-surgery ranges from 0.1% to 13% [4], reportedly lower for spontaneous leaks than other causes [5]. Prompt leak identification and appropriate intra-/postoperative management are critical in preventing such complications. In our cohort, complications were limited to one case of vision loss and one of pneumocephalus, both following surgery for ossifying fibroma. Another patient with a Glomus Jugulare tumor succumbed to intensive care unit-related comorbidities. Early intraoperative identification and immediate repair contributed significantly to overall favorable outcomes [8].
A consensus statement with the review of the literature identified high body mass index (BMI) as a significant risk factor for postoperative CSF leak [6]. Age and gender were not considered risk factors for CSF leaks. The risk of postoperative CSF leaks should be anticipated preoperatively and intraoperatively based on the intraoperative pressure of the leak, BMI, anatomical location of the defect, size of the defect, pathology of the lesion, and prior radiation exposure. Based on preoperative radiology, we categorized defect sizes as small, intermediate, and large when a skull base defect was identifiable with neuronavigation and intraoperative observation by the surgeon. In cases involving tumors or extensive cholesteatomas, the definitive size of the skull base defect could only be confirmed after removing the pathology. Appropriate reconstruction strategies are to be planned for a successful repair based on one or all of the factors mentioned above. None of our patients had a high BMI, and adequate reconstruction strategies were planned well in advance.
A thorough preoperative radiology assessment prevents untoward complications from CSF leaks. Prior skull base erosions, a low-lying olfactory groove, a steep inclination of the skull base, and intraoperative dural resections are potential risk factors that mandate an adequate reconstruction strategy.
An emphasis should be placed on the characteristics of the lesion we are dealing with. When a malignancy is concerned, complete removal with adequate resection margins is the principle of surgery. However, in benign cases such as ossifying fibroma, complete lesion removal is attempted along with removing its outer lamella. In cases where the outer lamella is irremovable due to its proximity to vital structures, the capsule should be drilled till the clear cortical bone is visible. One of our patients with juvenile ossifying fibroma developed a negative perception of light postoperatively, mainly due to the drill heat dissipating near the orbital apex. While neuronavigation confirmed tumor clearance and dural integrity, postoperative MRI revealed perineural edema around the optic nerve. Similar complications have been reported in the literature [9,10], emphasizing that extensive drilling near the optic canal and orbital apex poses a high risk of thermal or ischemic optic neuropathy. Preventive measures include meticulous irrigation during drilling, minimizing manipulation near the optic canal, and preoperative counselling of patients and families regarding the risk of vision loss in extensive skull base lesions. This case highlights the importance of balancing complete clearance of such lesions with preservation of critical neurovascular structures.
Another patient with ossifying fibroma developed a CSF leak during the attempt to remove the capsule near the skull base. Mild pneumocephalus was also seen in the postoperative radiology scan of this patient (Fig. 3). Large pneumocephalus can cause mass effect and neurological deterioration [11], but in our case, the patient remained clinically stable despite radiological worsening up to 4.5×5.1 cm. Conservative measures, including high-flow oxygen, bed head elevation, and lumbar drainage, resulted in gradual resolution by POD9. The accidental early removal of the lumbar drain did not adversely impact the outcome. This highlights that stable patients with radiological pneumocephalus may often be managed conservatively, provided close clinical and radiological monitoring is ensured. Fever in this patient was controlled with antibiotic escalation, and no meningitis or new neurological deficits developed. Our case emphasizes the importance of early postoperative imaging, vigilant follow-up, and a stepwise conservative management protocol for pneumocephalus after skull base reconstruction.
The reconstruction algorithm for skull base repair has been refined over the years [12,13]. A simple free mucosal flap will suffice for iatrogenic skull base defects, which are usually small. With larger defects and increasing complexity, generally arising after tumor excision, other modalities like autologous fat, temporalis fascia/fascia lata, muscle, cartilage/bone, DuraGen, oxidized cellulose, local/regional pedicled flaps, free flaps, fibrin glue, can be used in different combinations. While dealing with pathologies like a tumor, a complete resection with adequate margins should be targeted. This often leads to a larger-than-anticipated skull base defect [1]. A watertight seal to prevent any chances of pneumocephalus or ascending infection leading to meningitis is of paramount importance. In these cases, preventing postoperative complications and early ambulation also reduces the delay in administering postoperative radiotherapy wherever indicated.
The ideal management of temporal bone encephalocele is by fulguration or excision of herniating mass followed by multilayered closure of the bone defect in the tegmen with adequately sized chip from the outer cortical layer of cranium [7], cartilage [14,15], temporalis fascia, fat obliteration, muscle, and hydroxyapatite [16]. The three standard approaches for tegmen defects are transmastoid, middle fossa craniotomy and combined approaches. Wherein the pathology was a tumor necessitating an extensive resection, the skull base defect was repaired using the standard multilayered reconstruction along with muscle rotation and cul-de-sac closure. Postoperative immediate radiology is often indicated in pathologies such as malignancies, which may lead to raised intracranial pressure or dural collection post-resection and repair of skull base dura.
Although a few studies have shown that a lumbar drain insertion in skull base surgeries may reduce events of postoperative CSF leaks to less than 10%, especially in cases where the patients had raised BMI or the leak was found to be high flow intraoperatively [17], we did not follow the same as a routine in our patients, and only seven of our patients had postoperative lumbar drain placement. Lumbar drain placement has its bothersome complications, such as infection, pneumocephalus, spinal headache, nerve root damage or retained catheter. We recommend lumbar drain insertion in cases where the resection was extensive (such as skull base tumors and petrous cholesteatomas), causing a quite large defect or a very complex subsequent repair.
Skull base defects encountered in surgical practice can be further subclassified into iatrogenic, accidental, and planned surgical categories.
• Iatrogenic accidental defects are those unintentionally created during surgeries like endoscopic sinus surgery or mastoidectomy, often due to thin or dehiscent bone or aggressive instrumentation. These are frequently recognized intraoperatively by the presence of CSF leak and are repaired immediately.
• Planned surgical defects, on the other hand, are intentionally created during resections for malignant or invasive benign pathologies where the dura/skull base needs to be breached as part of oncologic clearance. These defects are anticipated, and preoperative planning allows for a structured reconstruction approach.
The surgical implications and outcomes of these two groups can vary. Accidental defects may present technical challenges due to their unexpected nature, but their confined and known boundaries allow for targeted repair. Conversely, planned iatrogenic defects often necessitate complex, multilayered reconstruction due to their larger size or involvement of critical neurovascular structures.
Iatrogenic defects—whether accidental or planned—differ significantly from those arising from other causes such as infections or trauma. One significant difference lies in the predictability of defect characteristics. In iatrogenic intraoperative defects, the surgeon has real-time knowledge of the site, size, and CSF flow dynamics, enabling immediate reconstruction. In non-iatrogenic defects, including post-infective, post-radiation, or post-tumor erosion, reconstruction planning must often rely on radiological assessment, which may not fully reveal the extent or complexity of the defect.
Furthermore, delayed presentations of iatrogenic CSF leaks—especially in revision surgeries like post-mastoidectomy—pose an added challenge due to loss of typical surgical landmarks, fibrosis, and altered tissue planes. These factors increase the difficulty of defect localization and reconstruction.
This study has important clinical implications for skull base reconstruction. In our series, all 26 reconstructions were successful with no secondary failures observed on follow-up. While this reflects the effectiveness of multilayered repair strategies combined with structured perioperative care, the limited sample size restricts the generalizability of this outcome. Moreover, two notable complications were encountered—one case of postoperative vision loss and one case of pneumocephalus—underscoring the inherent risks of skull base surgery, particularly during drilling near critical structures such as the orbital apex. These findings emphasize that although multilayered repair techniques are highly reliable in preventing CSF leaks, rare but serious complications must be anticipated and managed promptly.
Patient-specific strategies also played a key role—larger lateral skull base defects often required complex methods like muscle flaps or osteoplastic repair. In contrast, anterior defects were effectively managed endoscopically or with nasoseptal flaps. These insights support individualized reconstruction planning to optimize outcomes.
Finally, structured perioperative care—including selective lumbar drainage, head elevation, and acetazolamide—proved essential in minimizing complications and ensuring durable repair. These findings are particularly valuable for young skull base surgeons, offering practical guidance on tailoring reconstruction strategies and perioperative protocols for safer outcomes.
A stepwise skull base defect management protocol may be advised based on the experience from our center (Fig. 6).
Stepwise protocol for skull base defect management based on institutional experience. CT, computed tomography; MRI, magnetic resonance imaging; CSF, cerebrospinal fluid; ENT, otolaryngology.
This study has several limitations. Its retrospective design introduces potential selection and reporting bias, as data were dependent on the completeness of medical records and may not reflect uniform follow-up protocols. Being a single-center study, the findings may reflect institutional practices and surgical expertise that limit external generalizability. Additionally, the modest sample size may preclude detecting infrequent complications or subtle differences among reconstruction techniques. The absence of long-term follow-up restricts our ability to assess the durability of the multilayered repair, long-term functional outcomes, and delayed complications such as recurrent CSF leaks or flap necrosis. Future prospective, multicentric studies with standardized outcome measures and extended follow-up would help validate and expand these findings.
Skull base reconstruction is essential to prevent complications such as CSF leaks, meningitis, and pneumocephalus. In malignant cases, a watertight closure is critical for timely adjuvant therapy. With various autologous and allogenic materials now available, early intraoperative identification and tailored repair have significantly reduced complication rates. In our cohort, no secondary failures were observed, highlighting the effectiveness of structured reconstruction strategies. Successful outcomes also depended on thorough preoperative imaging and vigilant perioperative care.
Notes
Conflict of interest
No potential conflict of interest relevant to this article was reported.
Funding
None.
Ethical approval
The study was approved by the Institutional Ethics Committee of All India Institute of Medical Sciences (IEC No. AIIMS/IEC/2024/5286) and performed in accordance with the principles of the Declaration of Helsinki.
Patient consent
All patients, and in the case of minors their legal guardians, provided written informed consent for the publication and use of their images.
Author contributions
Conceptualization: all authors. Data curation: Palak Gupta, Srinivas Chadaram, Vidhu Sharma, Bikram Choudhury, Amit Goyal. Formal analysis: Palak Gupta, Srinivas Chadaram, Kapil Soni, Bikram Choudhury, Amit Goyal. Methodology: all authors. Visualization: Palak Gupta, Vidhu Sharma, Amit Goyal. Writing–original draft: Palak Gupta, Vidhu Sharma. Writing–review & editing: all authors. Investigation; Resources: Palak Gupta, Vidhu Sharma, Amit Goyal. Supervision: Srinivas Chadaram, Vidhu Sharma, Kapil Soni, Bikram Choudhury, Amit Goyal. Validation: Palak Gupta, Vidhu Sharma, Kapil Soni, Bikram Choudhury, Amit Goyal. All authors read and approved the final manuscript.
Abbreviations
BMI
body mass index
CSF
cerebrospinal fluid
CT
computed tomography
MRI
magnetic resonance imaging
POD
postoperative day