Delayed Cerebrospinal Fluid Leak Research Articles

Subfascial drains are safe and effective in preventing postoperative cerebrospinal fluid leaks after intradural spine tumor surgery.

Delayed cerebrospinal fluid (CSF) leaks are a known complication following intradural spinal tumor surgery. The placement of subfascial drains in these patients undergoing requisite intradural surgery is controversial. Here, we demonstrated that placing a subfascial drain on partial suction for 48 h, with early ambulation, proved to be safe and effective in preventing early/delayed recurrent CSF fistulas. Medical records of 17 patients undergoing surgery for intradural spinal tumors over a 30-month were reviewed. All patients underwent intradural tumor resection followed by primary dural closure, placement of Gelfoam in a non-compressive fashion, application of fibrin sealant, and utilization of a subfascial drain placed on partial suction for 48 h postoperatively. Patients are mobilized the morning following surgery. We tracked the incidence of postoperative recurrent CSF leaks, over drainage, infection, wound dehiscence, pseudo meningocele formation, and the reoperation rate. For the 17 patients, our programmed average utilization of subfascial drains was 48 h. Moreover, the average drain output was 165 mL. Over the 1-year follow-up period, no patient developed a recurrent early/ delayed CSF leak, there were no wound complications, nor need for revision surgery. Utilizing subfascial drains on partial suction following the resection of intradural spinal tumors with primary dural closure proved to be safe and effective.

Contributing factors for delayed postoperative cerebrospinal fluid leaks and suggested treatment algorithm.

Delayed postoperative cerebrospinal fluid (CSF) leaks are uncommon and largely unstudied complications. In this study we aim to identify their etiology and understand the efficacy of various reconstruction strategies. A retrospective chart analysis of 1017 endonasal skull base surgeries performed by a single neurosurgeon was completed identifying delayed CSF leaks (occurring >1 week after surgery). Seventeen cases of early (1-2 weeks after surgery) or delayed (>2 weeks after surgery) postoperative CSF leak were identified. The most common reconstruction during the initial surgery consisted of an inlay or gasket seal collagen matrix (82.4% of patients) with an onlay pedicled flap (76.5% of patients). Presenting symptoms of delayed CSF leak included rhinorrhea (82.4%), headache (41.2%), and meningitis (23.5%). The most common causes included flap dehiscence (17.6%); provoking events such as emesis, sneezing, or fall (17.6%); flap necrosis (11.8%); flap displacement (11.8%); and inadequate apposition of the flap, that is, folded flap (11.8%). Reconstructive techniques of the delayed CSF leak included fortification of the initial reconstruction with free fat grafts (29.4% of patients), combined collagen matrix with a fat graft (23.5% of patients), repositioning of the previous flap (11.8% of patients), and repair with a new flap (11.8% of patients). CSF diversion (spinal/ventricular drain or shunt) was used in 17.6% of patients. This study identifies the most common etiologic factors leading to a delayed CSF leak and its initial symptoms. Furthermore, it serves as the foundation for a reconstructive algorithm based on reinforcement of the initial repair with free abdominal fat graft with or without collagen matrix.

Orbitozygomatic Craniotomy via an Eyebrow Incision: Management of the Opened Frontal Sinus.

Background The eyebrow orbitozygomatic craniotomy is a minimally invasive approach that can access a wide variety of lesions. Unintentional breach of the frontal sinus frequently occurs and has been cited as a reason to avoid this approach. Lack of access to a large pericranial graft and the inability to completely cranialize the sinus requires alternate techniques of sinus repair. We describe a technique for repairing an opened frontal sinus and retrospectively reviewed complications related to this approach. Methods All patients, who underwent an orbitozygomatic craniotomy via an eyebrow incision by a single surgeon from August 1, 2012 to August 31, 2018, were included in this retrospective analysis. Data were collected on patient demographics, pathology treated, operative details, and perioperative morbidity. Follow-up ranged from 6 weeks to 6 years. Results Total 50 patients with a wide variety of pathologies underwent analysis. Frontal sinus breach occurred in 21 patients. All were repaired by the described technique. One patient (ruptured aneurysm) had a suspected cerebrospinal fluid (CSF) leak postoperatively that resolved without any additional intervention. One patient developed a pneumomeningocele 4 years postoperatively that required reoperation. No patient suffered any infection or delayed CSF leak. Conclusion Breach of the frontal sinus is common during eyebrow craniotomies. Despite reduced options for local repair, these patients have experienced no CSF leaks requiring intervention and no infections in our series. Long-term mucocele risk is not reliably determined with our length of follow-up. Breach of the frontal sinus is not a contraindication to the eyebrow approach.

Incidence & Management of Delayed Cerebrospinal Fluid Leaks after Lumbar Spinal Surgery-Analysis of 10 Cases

Objectives: Cerebrospinal fluid (CSF) leaks in degenerative lumbar spine surgery are common, however, delayed cerebrospinal fluid (CSF) leaks are quite rare in neurosurgical practice. Literature regarding its incidence and management is scant.Our aim was to describe the incidence & management of delayed CSF leaks after degenerative lumbar spine surgery. Material & Methods: This was a prospective study where all patients operated for lumbar disc or stenosis, who presented with the delayed CSF leak (> 1 week postoperatively) without intraoperative record of incidental durotomy were included. Data was collected about demographics, diagnosis, operative detail, postoperative course & management issues. Results: Ten out of 1128 patients developed delayed CSF leaks (0.89%). Mean age at the time of diagnosis was 52.1 ± 6.9 years with 6 (60%) males & 4 (40%) female. The most common spinal level was L5-S1 (50%). Eighty percent (n = 8) patients underwent primary surgery while 20% (n = 2) were revisions. Clinical features were headaches (80%), dizziness (70%) and altered sensorium in 20%. Mean time of the leak was 17.3 ± 2.2 days. Two patients resolved with bed rest and compression dressing while the lumbar drain was placed in 80%. Three (30%) patients of the 8 needed open repair of the dural defect. Complications of the CSF leak included wound infection in 60%, and meningitis in one (10%) patient. There were no cases of neurologic deficit. One case eventually developed infective discitis.

Delayed Cerebrospinal Fluid Leakage After Treatment of Skull Base Tumors: Case Series of 9 Patients

Cerebrospinal fluid (CSF) leaks usually happen in the early postoperative period. However, delayed CSF leak can happen several years after treatment. The purpose of this study was to review the characteristics and clinical course of delayed CSF leak after treating skull base tumors. We treated 9 patients with delayed CSF leak (occurring at least 3 months after treatment) between December 2015 and February 2018. Clinical data including initial treatment modality for skull base tumor, time between treatment and CSF rhinorrhea, and the result of endoscopic CSF repair were evaluated by retrospective chart review. Nine patients with delayed CSF leak were evaluated. The mean age was 42.3 ± 5.2 (mean ± SD) years. Six patients with pituitary adenomas, 1 with a Rathke cleft cyst, 1 with a meningioma, and 1 with an osteosarcoma were enrolled. Seven patients (78%) received radiation, and the mean radiation dose was 55.7 ± 2.6 Gy. The median time between the end of treatment for the skull base tumor and CSF rhinorrhea was 137.0 (interquartile range 24-145) months. Seven patients (78%) had meningitis at the time of operation. CSF rhinorrhea was treated with endoscopic reconstruction using a multilayer technique. Meningitis was improved without neurologic sequelae after reconstruction and antibiotic therapy. One patient had recurrence of CSF leak. Patients with delayed CSF leak are more likely to have a history of radiation and to present with meningitis at diagnosis. Therefore, CSF rhinorrhea should always be suspected, even several years after treatment, if early symptoms of CSF leak develop in patients with a history of skull base tumor.

A Multidisciplinary Technique Using Endonasal Endoscopic Approach with Intrathecal Fluorescin Under Neuronavigation Guidance in Patients with Cerebrospinal Fluid Leak

Background: Cerebrospinal fluid (CSF) is contained in an anatomic space, including cerebral ventricles and subarachnoid spaces and cisterns of brain and spine. Distraction of normal anatomy of any containing parts of this system will result in CSF leakage which could be associated with morbidity and mortality with variable degrees. In this study, we aimed to present 3 patients with history of blunt head trauma, who presented with delayed CSF leak and underwent endonasal endoscopic approach under image guided surgery using neuronavigation system with fluorescein illumination. Case Presentation: Three patients were referred to neurosurgery department of Shohada Tajrish Hospital with previous history of blunt head trauma, complained from CSF rhinorrhea. They underwent thin slice Computed Tomography scan and received intrathecal fluorescein for better visualization of leakage source under live endonasal endoscopic approach, using image guided neuronavigation technology. Autologous fat tissue and fascia was used to seal the leakage site. Clinical and imaging follow up at post-operative state as well as 1 st , 2 nd , 4 th and 8 th week revealed no post-operative complications and repeated CSF leakage. Conclusion: CSF rhinorrhea indicates abnormality in bony structure of skull base and is a major threat for ascending microbial infections and subsequent meningitis. Thus, identification of leakage site(s) and accurate surgical repair is necessary. We experienced a multidisciplinary approach which showed excellent results and no post-operative complications. Multidisciplinary approach with combination of endonasal endoscopic view under neuronavigation system improves accuracy and will minimize post-operative complications.

Cerebrospinal fluid rhinorrhea secondary to idiopathic intracranial hypertension: Long-term outcomes of endoscopic repairs.

Endoscopic endonasal repair of cerebrospinal fluid (CSF) rhinorrhea secondary to idiopathic intracranial hypertension (IIH), also known as pseudotumor cerebri or benign intracranial hypertension, has varying success rates, from 25-87%, with minimal morbidity. However, often these series have a relatively short-term follow-up. Given the pathophysiology of IIH, long-term follow-up is necessary to identify true CSF leak recurrence rates. Our investigation aimed to evaluate long-term outcomes in endoscopically repaired CSF leaks. A retrospective chart review of all the patients with CSF rhinorrhea due to IIH who met inclusion criteria between 1996 and 2009. Outcome measures included the following: demographics, intracranial pressure, location of skull base defect, presence of encephalocele and/or meningoencephalocele, surgical repair technique, treatment with acetazolamide, whether a ventriculoperitoneal shunt was inserted, location of recurrence, history of meningitis or previous sinus surgery, and duration of follow-up. Thirty-two patients with a total of 44 skull base defects were reviewed over a mean follow-up of 10.2 years. The mean body mass index and intracranial pressure were 36.8 kg/m(2) and 27.7 cm H2O, respectively. Seven patients (18%) required revision surgery at the same site or a distant site. We found no statistical significance that identified the recurrence risk in the outcome measurements most likely due to our small failure rate. However, early recurrences were noted to recur at the same repair site, whereas late recurrences were noted to recur at a distant site along the skull base. IIH is an increasingly recognized entity treated by otorhinolaryngologists. We present the first long-term IIH CSF leak repair series. Long-term follow-up is necessary because delayed CSF leaks occur in this population.

Delayed CSF Rhinorrhea after Gamma Knife Radiosurgery with or Without Preceding Transsphenoidal Hypophysectomy for Pituitary Pathology

Background: Spontaneous skull base cerebrospinal fluid (CSF) leak after gamma knife radiosurgery (GKRS) is a very rare complication, with only 5 cases reported in patients who did not also undergo prior surgery. In patients who were treated with both GKRS and transsphenoidal hypophysectomy (TSSH) for pituitary lesions, early CSF leak occurs at a comparable rate to the general TSSH population (4%), typically within weeks-to-months of the primary operation. Delayed CSF leak occurring more than a year after TSSH, GKRS, or dual therapy, is exceedingly rare.

Sellar Reconstruction and Rates of Delayed Cerebrospinal Fluid Leak after Endoscopic Pituitary Surgery.

Objectives Delayed cerebrospinal fluid (CSF) leaks are a complication in transsphenoidal surgery, potentially causing morbidity and longer hospital stays. Sella reconstruction can limit this complication, but is it necessary in all patients? Design Retrospective review. Setting Single-surgeon team (2005-2012) addresses this trend toward graded reconstruction. Participants A total of 264 consecutive patients with pituitary adenomas underwent endoscopic transsphenoidal resections. Sellar defects sizable to accommodate a fat graft were reconstructed. Main outcomes Delayed CSF leak and autograft harvesting. Results Overall, 235 (89%) had reconstruction with autograft (abdominal fat, septal bone/cartilage) and biological glue. Delayed CSF leak was 1.9%: 1.7%, and 3.4% for reconstructed and nonreconstructed sellar defects, respectively (p = 0.44). Complications included one reoperation for leak, two developed meningitis, and autograft harvesting resulted in abdominal hematoma in 0.9% and wound infection in 0.4%. Conclusion In our patients, delayed CSF leaks likely resulted from missed intraoperative CSF leaks or postoperative changes. Universal sellar reconstruction can preemptively treat missed leaks and provide a barrier for postoperative changes. When delayed CSF leaks occurred, sellar reconstruction often allowed for conservative treatment (i.e., lumbar drain) without repeat surgery. We found universal reconstruction provides a low risk of delayed CSF leak with minimal complications.

Double flap technique for reconstruction of anterior skull base defects after craniofacial tumor resection: technical note

Successful reconstruction of large anterior skull base (ASB) defects after craniofacial resection of malignant skull base tumors is paramount for preventing cerebrospinal fluid (CSF) fistulas. The vascularized pedicled pericranial flap (PCF) has been the gold standard for repairing ASB defects after transbasal transcranial approaches. However, flap necrosis and delayed CSF leaks can occur after adjuvant radiation therapy. We describe a "double flap" reconstruction technique in which the PCF is augmented inferiorly by a secondary vascularized pedicled nasoseptal flap (NSF) that is harvested and rotated using an endoscopic endonasal approach. This technique is illustrated in 2 patients who underwent a combined cranionasal (transbasal and endoscopic endonasal) approach for large sinonasal malignancies with significant intracranial extension (1 esthesioneuroblastoma, 1 sinonasal teratocarcinosarcoma). After tumor removal via a combined cranionasal approach, primary repair of the ASB dural defect was performed with a free patch graft. The ASB defect was then repaired using the double flap technique with a vascularized PCF from above and augmented with a vascularized NSF from below. Postoperatively, there were no complications of CSF leakage, meningitis, or tension pneumocephalus in both patients. After subsequent radiation therapy, the double flap repair remained intact at 2 years postoperatively in both patients. The double flap skull base reconstruction technique provides an additional barrier of vascularized tissue to prevent CSF leakage, meningitis, tension pneumocephalus, and postradiation necrosis. This technique is a viable option if a combined transcranial and transnasal endoscopic tumor resection is performed and postoperative radiation is anticipated.

Novel Harvesting Technique of the Nasoseptal Flap in Large Sphenoidotomy Undergoing Revision With Transsphenoidal Approach

Blood supply to the nasoseptal flap (NSF) can be disrupted by large sphenoidotomies in an endonasal transsphenoidal approach (TSA). In such patients with recurrent sellar tumor, an NSF can be unavailable in the sellar reconstruction. Herein, we present a new harvesting method of the NSF in revision endonasal TSA, where injury to its vascular pedicle of the NSF by a large sphenoidotomy is highly suspected. Data for patients with large sphenoidotomy who underwent revision endonasal TSA and NSF at Chonnam National University Hwasun Hospital were collected consecutively from January 2010 to July 2011. A retrospective review of these cases was performed. Seven patients with a previous large sphenoidotomy requiring revision TSA underwent a new modification technique of the NSF. Pathologies included 6 pituitary adenomas and 1 craniopharyngioma. All patients had intraoperative cerebrospinal fluid (CSF) leak, necessitating multilayered sellar reconstruction, including NSF. No patients had flap necrosis, and closure of CSF leak was possible in 6 of 7 patients. Delayed CSF leak in a patient was due to the migration of the flap over the defect, and no septal complications such as septal perforation and saddle nose were noted. Our harvesting technique of the NSF in patients with recurrent sellar tumors is reliable for sellar reconstruction where the NSF may be unavailable because of previous large sphenoidotomy. This novel technique has an excellent success rate in flap viability and minimal additional endonasal morbidity.

A management algorithm for cerebrospinal fluid leak associated with anterior skull base fractures: detailed clinical and radiological follow-up

Detailed outcome data for the management of anterior skull base fractures associated with cerebrospinal fluid (CSF) leakage is lacking. We present detailed follow-up data of a single-center study using a predetermined algorithm for the management of CSF leakage secondary to traumatic fractures. A number of 138 consecutive patients were included in the analysis; all patients underwent high-resolution computed tomography (CT) scanning at time of admission with β(2)-transferrin testing used to confirm CSF leakage. Patients with acute surgical indications were operated as emergent; leaks were repaired at the time of initial surgery in patients with intracranial pressure < 15cm H(2)O. The remainder of the study population was managed conservatively including use of prophylactic antibiotics; lumbar drainage (LD) catheters were placed in those patients with leakage persisting beyond 48h. Leaks lasting longer than 5days underwent microsurgical repair using an intradural bicoronal approach. One-year follow-up assessment included evaluation of neurological status, Glasgow Outcome Scale (GOS), and repeat head CT. Twenty eight patients (26.9%) underwent emergent surgery, 15 of whom had simultaneous CSF leak repair, whereas 76 patients (73.1%) underwent delayed CSF leak repair between days 5 and 14. Postoperative meningitis rate was low (1.9%). Postoperative CSF leak (1.9%) was managed by intradural or transnasal endoscopic operation. Comparable rates of anosmia and frontal lobe hypodensities were seen in the surgical and conservatively managed subgroups. The presented algorithm, utilizing prophylactic antibiotics, trial of LD, acute and/or delayed intradural microsurgery, yields favorable outcomes. Large randomized controlled trials are needed to better define the role of prophylactic antibiotics and to better characterize the optimal timing and approach of surgical repair.

Double Flap Technique for Reconstruction of Anterior Skull Base Defects after Craniofacial Tumor Resection

Background: Successful reconstruction of large anterior skull base (ASB) defects after craniofacial resection of malignant skull base tumors is paramount for preventing cerebrospinal fluid (CSF) fistulas. The vascularized pedicled pericranial flap has been the gold standard for repair. However, flap necrosis and delayed CSF leaks can occur after adjuvant radiation therapy. We describe a double flap repair, where the pericranial flap is augmented inferiorly by a secondary vascularized nasoseptal flap (NSF) that is harvested and rotated using an endonasal endoscopic approach. This technique is illustrated in two patients who underwent resection of large sinonasal/anterior skull base malignancies with significant intracranial extension.

Diagnosis and treatment of traumatic delayed cerebrospinal fluid rhinorrhea

To explore the etiological factor, diagnostic localization and treatment of delayed cerebrospinal fluid (CSF) rhinorrhea. The medical records of 79 patients who had undergone endoscopic repair of CSF rhinorrhea between 2000 and 2008 were reviewed. Thirteen patients with CSF leaks occurred 3 months after head trauma. All 13 patients with delayed CSF leak were retrospectively evaluated with CT or MRI and surgically treated. The operative findings were compared with the results of CT or MRI to estimate the diagnostic value of imaging technique. Bony defects had been found on CT scanning in all 13 patients. Neural tissue herniation into the nasal sinuses was found in 11 patients during the surgery. The sizes of the leak ranged from 0.1 cm x 0.2 cm to 1.2 cm x 1.5 cm. Reconstruction of the skull base was done through endoscopic approach. No complications were found. Thirteen patients were followed up from 12 - 36 months, and none was recurrence. Herniation of mucosal tissue into the nasal sinuses after skull base defects could result in delayed CSF leak. CT and MRI can clearly show the skull base defects and neural tissue herniation. Endoscopic closure of CSF leaks was both safe and effective.

Endoscopic Management of Idiopathic Spontaneous Skull Base Fistula Through the Clivus

Cerebrospinal fluid (CSF) is produced by the choroid plexus of the lateral third and fourth ventricles at a rate of 0.35 mL/min. The CSF flows into the subarachnoid space and is absorbed by the arachnoid villi in the sagittal sinus. Normal CSF pressure is 5 to 15 cm H2O. Neurologic symptoms may occur when CSF pressure reaches more than 15 to 20 cm H2O. 1 Three concomitant factors are needed for CSF leakage: an osseous defect, a meningeal disruption, and a pressure gradient. 2 Cerebrospinal fluid rhinorrhea may be classified as a function of its site, its cause, and the intensity of intracranial pressure. Nontraumatic fistulas with normal intracranial pressure (4% of all fistulas) constitute a challenge because of their unclear pathophysiologic features. 3 Some authors have used the termspontaneous to include CSF leaks due to a tumor, a delayed CSF leak after head trauma, or CSF leaks associated with congenital malformations of the skull base. Currently, the term spontaneous is associated with CSF leaks without any known cause. For those reasons, it seems preferable to use the term idiopathic spontaneous CSF leak. 4 Some risk factors of idiopathic, spontaneous CSF rhinorrhea in the sphenoid sinus have been described in the literature: skull base malformation; overpneumatized sphenoid sinus, particularly with a lateral extension (present in 16%-27% of adults); empty sella syndrome; and obesity. Obesity is associated with an increase of abdominal and thoracic pressures, leading to cranial venous congestion and permanent benign intracranial hypertension. 5-7

Sellar floor reconstruction after transsphenoidal surgery using fibrin glue without grafting or implants: technical note

Background Different techniques have already been described for reconstructing the sellar floor after transsphenoidal (TS) procedures. This paper reports on the use of fibrin glue alone without grafting or the use of implants in the reconstruction of the sellar floor after TS. Methods Five hundred sixty-seven patients who submitted to TS for pituitary and sellar region tumors were studied. No intraoperative cerebrospinal fluid (CSF) leak occurred in 503 patients (group 1); in the remaining 64 patients (group 2), intraoperative CSF leak was noted. In group 1 patients, closure of the sellar floor consisted of packing the surgical bed with hemostatic material only. When CSF leak was noted, the surgical bed was covered with a layer of hemostatic material and the intrasellar space was filled up with fibrin glue. An additional layer of hemostatic material was added at the topography of the preexisting sellar floor, and a second amount of fibrin glue was applied over it. At the end of surgery, a continuous lumbar CSF drainage system was installed in group 2 patients and kept for 5 days. Prophylactic antibiotics were administered during this period. Results We did not observe delayed CSF leak, meningitis, or visual loss in group 1 patients. In group 2, 2 patients presented with complications: 1 patient got meningitis but no overt CSF leak, and the other disclosed a delayed postoperative leak treated by reoperation. Discussion Our results showed that closure of the sellar floor with hemostatic material and fibrin glue without grafting or the use of implants is a safe and efficient method to prevent postoperative complications after TS. Generally speaking, there is no need for grafting or the use of implants at the end of TS.

Risk factors of cerebrospinal fluid rhinorrhea following transsphenoidal surgery

Cerebrospinal fluid (CSF) rhinorrhea is a potentially life-threatening complication following transsphenoidal surgery (TSS). To elucidate the risk factors that may affect the incidence of postoperative CSF rhinorrhea, we retrospectively reviewed 200 consecutive cases of TSS performed by a single surgeon for 168 adenomas and 32 other sellar and parasellar lesions. Intra-operative CSF leakage was encountered in 38 cases (19.0%). Its incidence did not correlate to tumor size. Among 4 microadenoma cases with CSF leak, 3 were ACTH adenomas. In contrast, postoperative CSF rhinorrhea was observed in 5 cases (2.5%), all following TSS for adenomas. It was frequently noted in cases with prior TSS (3/40, 7.5%, p = 0.0235) and prior radiotherapy (2/7, 28.6%, p < 0.0001). Two cases who required surgical intervention had received TSS and radiotherapy previously. The risk of postoperative CSF rhinorrhea is significantly increased in cases with prior TSS or radiotherapy or both. In addition, these cases tended to show delayed CSF leaks and require sellar reconstruction for its treatment. When a CSF leak was encountered during TSS in these high-risk cases, thorough sellar reconstruction and long-term follow-up is necessary.

Persistent posttraumatic cerebrospinal fluid leakage

Persistent posttraumatic cerebrospinal fluid (CSF) leakage frequently complicates skull base fractures. Although many CSF leaks will cease without treatment, patients with CSF leaks that persist greater than 24 hours may be at increased risk for meningitis, and many will require surgical intervention. The authors reviewed their 15-year experience with posttraumatic CSF leaks that persisted longer than 24 hours. The authors reviewed the medical records of 51 patients treated between 1984 and 1998 with CSF leaks that persisted for 24 hours or longer after traumatic head injury. In 27 patients (55%) spontaneous resolution of CSF leakage occurred at an average of 5 days posttrauma. In 23 patients (45%) surgery was required to resolve the leakage. Eight patients (16%) with occult CSF leaks presented with recurrent meningitis at an average of 6.5 years posttrauma. Forty-three (84%) patients with CSF leaks sustained a skull fracture, most commonly involving the frontal sinus, whereas parenchymal brain injury or extraaxial hematoma was demonstrated in only 18 patients (35%). Delayed CSF leaks, with an average onset of 13 days posttrauma, were observed in eight patients (16%). Among patients with clinically evident CSF leakage, the frequency of meningitis was 10% with antibiotic prophylaxis, and 21% without antibiotic prophylaxis. Thus, prophylactic antibiotic administration halved the risk of meningitis. A variety of surgical approaches was used, and no significant neurological morbidity occurred. Three (13%) of 23 surgically treated patients required additional surgery to treat continued CSF leakage. A significant proportion of patients with CSF leaks that persist greater than 24 hours will require surgical intervention. Prophylactic antibiotic therapy may be effective in this group of patients. Patients with skull base or frontal sinus fractures should be followed to detect the occurrence of delayed leakage. Surgery-related outcome is excellent.