Abstract
Objective Diagnosis of cerebrospinal fluid (CSF) leaks is sometimes challenging in the postoperative period following pituitary and ventral skull base surgery. Intrathecal fluorescein (ITF) may be useful in this setting.
Design Retrospective chart review.
Setting Tertiary care center.
Methods and Participants All patients who underwent pituitary and ventral skull base surgery performed by a single rhinologist between January 2017 and March 2020 were included. There were 103 patients identified. Eighteen patients received 20 ITF injections due to clinical suspicion for CSF leak during the postoperative period without florid CSF rhinorrhea on clinical exam. Computed tomography scans with new or increasing intracranial air and intraoperative findings were used to confirm CSF leaks. Clinical courses were reviewed for at least 6 months after initial concern for leak as the final determinate of CSF leak.
Main Outcome Measures Specificity and safety of ITF.
Results Eleven (61%) ITF patients were female and 7 (39%) were male. Average patient age was 52.50 ± 11.89. There were six patients with confirmed postoperative CSF leaks, 3 of whom had evaluations with ITF. ITF use resulted in 2 true positives, 1 false negative, 17 true negatives, and 0 false positives. ITF sensitivity was 67%, specificity was 100%, and positive and negative predictive values were 100 and 94.4%, respectively. There were no adverse effects from ITF use.
Conclusions Existing modalities for detecting postoperative CSF leaks suffer from suboptimal sensitivity and specificity, delayed result reporting, or limited availability. ITF represents a specific and safe test with potential utility in the postoperative setting.
Keywords: cerebrospinal fluid leak, fluorescein, postoperative period, diagnostic tests
Introduction
Postoperative cerebrospinal fluid (CSF) leaks following pituitary and ventral skull base surgery carry a significant risk of morbidity, including meningitis, tension pneumocephalus, and brain herniation. 1 2 Postoperative CSF leak rates after skull base surgery vary based on the location of the defect, tumor size, and patient factors. 3 Additionally, after extensive sinonasal dissection, as well as the use of packing and splints, patients commonly report nonspecific symptoms including anterior and posterior rhinorrhea, which may or may not represent a CSF leak. Postoperative CSF leaks can vary significantly in presentation from obvious florid rhinorrhea to a more occult, intermittent, and less copious rhinorrhea. These factors combine to make the prompt and definitive diagnosis of CSF leak challenging in the postoperative setting in some of the cases, and the workup of these symptoms may result in extensive diagnostic tests, prolonged hospital stays, and unnecessary return trips to the operating room (OR). 2 4 Intrathecal fluorescein (ITF) injection is a diagnostic modality that may aid in accurate CSF leak diagnosis. Previous studies have examined ITF as part of the diagnostic workup of spontaneous CSF leaks; however, little is known about the utility of ITF in the postoperative period. 2 4 5 At present, postoperative CSF leaks are diagnosed at most institutions using a variable combination of patient reporting, radiographic imaging modalities, beta-2 transferrin testing, and bedside endoscopic evaluation with and without the use of ITF injection. 6
ITF administration carries with it the risk of chemical meningitis, but the low doses of ITF used for the diagnosis of CSF leaks have been well established as having an excellent safety profile. 5 7 Commonly reported adverse effects experienced by patients following ITF administration are transient, nonspecific, and frequently attributable to the lumbar puncture/drain rather than fluorescein itself. 7
This study analyzed the utility of ITF injection as a screening and diagnostic tool for CSF leaks in the postoperative period following pituitary and ventral skull base surgery for patients without florid CSF rhinorrhea on clinical exam.
Materials and Methods
Data Collection
The current study was approved by the Institutional Review Board at the New York University Langone Health. A retrospective chart review was conducted on all patients who underwent pituitary or ventral skull base surgery performed by a single rhinologist (SL) between 1/1/2017 and 4/30/2020 at our tertiary care center. From this sample, patients were identified who underwent ITF injection in the postoperative period due to clinical suspicion for the presence of a CSF leak, but without florid CSF rhinorrhea. These patients fell into two categories—those who already had a planned lumbar drain that was placed intraoperatively, and those who had either a lumbar drain or lumbar puncture performed in the postoperative period to introduce the ITF. Data were collected on patient demographics, comorbidities (including body mass index [BMI], diabetes mellitus, hypertension, and obstructive sleep apnea [OSA]), primary versus revision surgery, the presence of an intraoperative CSF leak, placement and timing of lumbar drain/puncture, and the indication for surgery (e.g., pituitary adenoma, meningioma).
Endoscopic Fluorescein Test
After undergoing skull base surgery, patients with postoperative rhinorrhea deemed by the senior author and/or neurosurgeon to be clinically concerning for possible CSF leak were injected via either lumbar drain or lumbar puncture with 0.1 to 0.2 mL of fluorescein (at 500mg/5mL per vial, sourced from the inpatient pharmacy at our institution) in 10mL of sterile CSF or sterile saline, for a cumulative dose of 10 to 20 mg of fluorescein. If the patient already had a lumbar drain that had been placed intraoperatively, then this was utilized. If the patient did not have intrathecal access, then either a lumbar drain or lumbar puncture was placed to instill the fluorescein. Patients were instructed to lie flat for 60 minutes after placement of the lumbar drain or puncture. Lumbar drains were clamped after administration of the fluorescein. Variations in fluorescein dosage were due to differences in the practice of the neurosurgeon performing the injection. All patients were premedicated with a one-time dose of 1 g levetiracetam, 10 mg dexamethasone, and 25 mg diphenhydramine. Patients were then evaluated for CSF leak 2 to 12 hours after ITF injection via bedside flexible nasal endoscopy of the surgical field and nasopharynx using standard white light (with the exception of one patient who preferred to be examined under general anesthesia and one patient who was examined under general anesthesia for the combined reasons of possible CSF leak and possible persistent hormone-secreting pituitary adenoma, as detailed in the results section). Prior to endoscopy, cottonoids soaked in 4% topical lidocaine and oxymetazoline were placed into patients' bilateral nasal cavities for at least 5 minutes to achieve appropriate anesthesia, and patients were positioned sitting upright with their chin touching their chest for 2 to 5 minutes to increase the likelihood of detecting a small leak. Valsalva maneuver was not performed prior to endoscopy. In addition, all gauze or tissue that was used to wipe any nasal secretions was kept at bedside and routinely inspected for any signs of fluorescein. A combination of computed tomography (CT) scans and intraoperative findings (in patients who required takeback to the OR) was used at the time of treatment as secondary confirmation for CSF leak detection. Patients' ultimate clinical courses were also reviewed for at least 6 months following initial concern for CSF leak, and this outcome was used as the final determination of CSF leak for this study. Because the goal of this study was to determine the utility of ITF as a diagnostic tool in the postoperative period for equivocal cases of CSF leak, those patients who had definitive evidence of postoperative CSF leak on clinical exam (i.e., florid CSF rhinorrhea) were not considered for the ITF group during analysis, even if they underwent ITF administration to aid in intraoperative CSF leak localization.
Statistical Analysis
All statistical analyses were performed using IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, New York, United States). Sensitivity of ITF injection was calculated as true positives/(true positives + false negatives); specificity was calculated as true negatives/(true negatives + false positives); the positive predictive value (PPV) was calculated as true positives/(true positives + false positives); and the negative predictive value (NPV) was calculated as true negatives/(true negatives + false negatives). A Clopper-Pearson interval, as previously described in the literature, was used to determine the confidence intervals for sensitivity, specificity, PPV, and NPV. 8 A Fischer's exact test was used to determine the relationship between postoperative infection and patient comorbidities and perioperative events. For significant associations, a binary logistic regression analysis was used to determine that was true independent risk factors for postoperative infection. All p -values reported were two-sided; statistical significance was defined as p < 0.05.
Results
From 1/1/2017 to 4/30/2020, 103 patients underwent pituitary or ventral skull base surgery performed by the senior author ( Table 1 ). Sixty-three of these patients were female (61%), with an average age of 51.83 ± 15.94 years and an average BMI of 29.83 ± 6.70. Fifty patients (49%) had hypertension, 19 (18%) had diabetes, and 19 (18%) had OSA. The most common indication for surgery was pituitary adenoma ( n = 63, 61%), followed by meningioma ( n = 8, 8%), encephalocele ( n = 8, 8%), spontaneous CSF leak ( n = 5, 5%), and Rathke's cleft cyst ( n = 4, 4%).
Table 1. Demographic information for the study population.
| n | % of cohort | ||
|---|---|---|---|
| Total population | 103 | 100 | |
| Gender | Male | 40 | 38.83 |
| Female | 63 | 61.17 | |
| Average age | 51.83 ± 15.94 | – | |
| Average BMI | 29.83 ± 6.70 | – | |
| Comorbidities | Diabetes | 19 | 18.45 |
| Hypertension | 50 | 48.54 | |
| Obstructive sleep apnea | 19 | 18.45 | |
| Primary versus revision surgery | Primary surgery | 82 | 79.61 |
| Revision surgery | 21 | 20.39 | |
| Indication for surgery | Pituitary adenoma | 63 | 61.17 |
| Meningioma | 8 | 7.77 | |
| Encephalocele | 8 | 7.77 | |
| Spontaneous | 5 | 4.85 | |
| Rathke's cleft cyst | 4 | 3.88 | |
| Craniopharyngioma | 4 | 3.88 | |
| Squamous cell carcinoma | 2 | 1.94 | |
| Trauma | 1 | 0.97 | |
| Esthesioneuroblastoma | 1 | 0.97 | |
| IgG4 related pachymeningitis | 1 | 0.97 | |
| Mucocele | 1 | 0.97 | |
| Chronic rhinosinusitis with nasal polyposis | 1 | 0.97 | |
| Carcinoid | 1 | 0.97 | |
| Cavernoma | 1 | 0.97 | |
| Pyogenic granuloma | 1 | 0.97 | |
| Adenoid cystic carcinoma | 1 | 0.97 |
Abbreviation: BMI, body mass index.
Of these 103 patients, there were a total of 6 (5.8%) patients who had postoperative CSF leaks. A total of 18 patients (17%) were injected with fluorescein a total of 20 times to evaluate for the possibility of a postoperative CSF leak ( Table 2 ). Eleven (61%) of these patients were female and 7 (39%) were male. Average age was 52.50 ± 11.89 years and average BMI was 32.97 ± 6.90. Twelve (67%) had hypertension, 9 (50%) had OSA, and 3 (17%) had diabetes. The most common etiology was pituitary adenoma ( n = 9, 50%), followed by meningioma ( n = 3, 17%) and encephalocele ( n = 2, 10%). Among the 18 patients who underwent ITF injection, 15 had a documented CSF leak during the initial procedure. Lumbar drains were placed intraoperatively in 9 of the 18 patients, while either a lumbar drain or lumbar puncture was performed in the remaining 9 patients in the postoperative period to administer ITF. Of the 9 patients who did not receive intraoperative lumbar drains, 4 received postoperative lumbar drains, and 5 received lumbar punctures for ITF administration. ITF injection was performed a median of 5.5 (interquartile range: 2.75) days after initial procedure.
Table 2. Demographic information for patients injected with fluorescein.
| n | % of cohort | ||
|---|---|---|---|
| Total population | 18 | 100 | |
| Gender | Male | 7 | 38.89 |
| Female | 11 | 61.11 | |
| Average age | 52.50 ± 11.89 | – | |
| Average BMI | 32.97 ± 6.90 | – | |
| Comorbidities | Diabetes | 3 | 16.67 |
| Hypertension | 12 | 66.67 | |
| Obstructive sleep apnea | 9 | 50.00 | |
| Primary versus revision surgery | Primary surgery | 12 | 66.67 |
| Revision surgery | 6 | 33.33 | |
| Indication for surgery | Pituitary adenoma | 9 | 50.00 |
| Meningioma | 3 | 16.67 | |
| Encephalocele | 1 | 5.56 | |
| Spontaneous | 1 | 5.56 | |
| Rathke's cleft cyst | 1 | 5.56 | |
| Craniopharyngioma | 1 | 5.56 | |
| Trauma | 1 | 5.56 | |
| Cavernoma | 1 | 5.56 |
Abbreviation: BMI, body mass index.
The average BMI of patients who underwent ITF injection (32.97 ± 6.90) was significantly higher than those patients who did not receive ITF (29.17 ± 6.463), ( p = 0.043, 95% confidence interval [CI]: 7.548–-0.0137). However, BMI was not associated with proven CSF leak in this cohort: average BMI in patients who experienced a CSF leak was 32.54 ± 3.726 versus 29.67 ± 6.784 in patients who did not experience a CSF leak ( p = 0.129, 95% CI: 1.06–6.78). OSA was also more prevalent in those who received ITF: 56% of ITF patients had OSA versus only 11% of non-ITF patients ( p < 0.001). The presence of OSA was associated with CSF leak on univariate analysis ( p = 0.019). There were no other significant demographic differences between groups.
Three patients received multiple injections of ITF. One patient was discharged in good condition following repair of a spontaneous CSF leak and presented a year later with a recurrent encephalocele requiring surgery. ITF was used during both admissions postoperatively to assess for a suspected leak. One patient developed a clinically obvious CSF leak on postoperative day 4 after pituitary macroadenoma resection and underwent CSF leak repair with intraoperative ITF injection for localization of the leak (this ITF administration did not fit criteria to be analyzed in the current study). After repair, there was concern for persistent leak, and the patient underwent bedside ITF injection, which was negative. The patient was subsequently taken to the OR after development of a clinically apparent leak (this patient was considered to have a false negative postoperative ITF test). One patient underwent resection of a pituitary macroadenoma, was discharged, presented to the emergency department 1 day later for suspected leak, and had lumbar drain placement with ITF injection which was negative. The patient was discharged, and then presented again 18 days later for suspected leak. The patient was offered bedside lumbar drain placement and ITF injection but declined, preferring to receive general anesthesia for lumbar drain placement and ITF administration. The patient was examined in the OR after ITF administration and again was determined to not have a leak. One other patient included in the cohort underwent ITF injection and examination under anesthesia. This patient was not examined at the bedside because in addition to a suspected CSF leak there was concern on imaging for possible residual hormone secreting pituitary adenoma and it was determined that an OR trip could address both possibilities simultaneously. No CSF leak was detected in this patient.
There were 8 total instances of post-operative CSF leak among 6 patients. One patient never received ITF, and one patient received fluorescein but was not considered in the ITF group on analysis because the CSF leak was obvious on conventional clinical exam and ITF was only administered perioperatively to help with localization of the leak during operative repair. Of these six patients, all except one (83%) patient had a documented CSF leak during their initial procedure. Table 3 describes the 6 patients who experienced postoperative CSF leaks.
Table 3. Pathologic information, hospital course, and follow-up for the 6 patients who developed a postoperative CSF leak during the study period.
| Patient | Pathology | Intraoperative leak? | Repair technique (primary operation) | Time to postoperative leak (days) | Method of repair | Time from leak to last appt (months) |
|---|---|---|---|---|---|---|
| 1 | Craniopharyngioma | Yes | AlloDerm, nasoseptal flap | 12 | Fascia lata, Nasoseptal flap | 28 |
| 2 | Meningioma | Yes | Duragen, nasoseptal flap | 19 | AlloDerm, nasoseptal flap, NasoPore, merocel | 30 |
| 3 | Rathke's cleft cyst | No | Free mucosal graft | 60 | AlloDerm, nasoseptal flap, NasoPore, Merocel | 27 |
| 4 | Craniopharyngioma | Yes | AlloDerm, nasoseptal flap | 3 | AlloDerm, nasoseptal flap, Merocel | 37 |
| 5 | Macroadenoma | Yes | AlloDerm, nasoseptal flap | 9 | AlloDerm, nasoseptal flap, NasoPore, Merocel | 8 |
| 6 | Macroadenoma | Yes | AlloDerm, nasoseptal flap, Nasopore | 7 | AlloDerm, nasoseptal flap, NasoPore, Merocel | 18 |
Abbreviation: CSF, cerebrospinal fluid.
CT scans with new or increasing intracranial air and intraoperative findings were both used at the time of treatment to confirm leaks in all eight cases of postoperative CSF leak. Patient charts were also subsequently reviewed for at least 6 months after initial concern for CSF leak to ensure that no CSF leaks or bouts of meningitis were missed by these detection modalities that later presented clinically. Using these criteria, no CSF leaks were missed. ITF injection in this cohort was found to have a sensitivity of 0.667 (95% CI: 0.094–0.992) and a specificity of 1.000 (95% CI: 0.805–1.000). The PPV for the test was 1.000 (95% CI: 0.158–1.000), and the NPV was 0.944 (95% CI: 0.727–0.999; Table 4 ).
Table 4. Statistical values.
| Fluorescein injection | Value | 95% CI |
|---|---|---|
| Sensitivity | 0.667 | 0.094–0.992 |
| Specificity | 1.00 | 0.805–1.000 |
| Positive predictive value | 1.00 | 0.158–1.000 |
| Negative predictive value | 0.944 | 0.727–0.999 |
Abbreviation: CI, confidence interval.
Of the 103 patients, there were 5 cases of meningitis or presumed meningitis. Three patients who received ITF developed meningitis at some point during their postoperative course: one patient had methicillin-susceptible Staphylococcus aureus isolated in CSF culture, one patient had Streptococcus mitis/Streptococcus oralis isolated, and one patient presented to an outside hospital with meningitis was started on broad-spectrum antibiotics, and was transferred to our center. No organism was isolated, and this patient later received ITF injection to diagnosis a CSF leak. One patient who did not receive ITF developed meningitis, with Klebsiella pneumoniae isolated in CSF culture. One patient was started on broad-spectrum antibiotics for possible meningitis, but this was stopped after 4 days due to low clinical suspicion for infection.
Receiving ITF was associated with postoperative meningitis on univariate analysis ( p = 0.027). Also associated with meningitis were comorbid OSA ( p = 0.003), placement of an intraoperative lumbar drain ( p = 0.024), proven postoperative CSF leak ( p = 0.001), reoperation ( p = 0.005), and placement of a postoperative lumbar drain ( p = 0.018). Lumbar puncture was not associated with postoperative meningitis on univariate analysis ( p = 0.269). Binary logistic regression demonstrated that undergoing fluorescein injection was not an independent risk factor for postoperative meningitis ( p = 0.771). There were no complications or adverse reactions from the injection of fluorescein itself.
Discussion
Postoperative CSF leaks are one of the most common complications of pituitary and ventral skull base surgery and are associated with multiple potentially dangerous sequelae including meningitis and pneumocephalus. 1 2 Rapidly and accurately diagnosing CSF leaks in the postoperative setting is sometimes challenging as it is common for patients to have significant anterior and posterior rhinorrhea after sinonasal dissection. Additionally, nasal packing placed during surgery often obstructs the drainage of nasal and lacrimal secretions into the oropharynx, thereby leading to rhinorrhea that may easily be mistaken for a CSF leak.
Multiple methods have been proposed to aid in the bedside identification of CSF leaks, including use of the “ring sign” and glucose testing; however, the existing literature does not support the use of these tests due to their low accuracy and high potential for misdiagnosis. 9 Imaging studies such as noninvasive and invasive magnetic resonance cisternography, and CT cisternography may not accurately diagnose a leak if not actively leaking at the time of the study. Beta-2 transferrin has been demonstrated to be a valuable confirmatory test for CSF rhinorrhea, but the results of this test typically take at least several days to return, and collecting adequate specimen volume for this test can be challenging in postoperative patients with intermittent rhinorrhea. 9 At our institution, beta-2 transferrin testing takes between 4 and 7 days to result, reducing its clinical utility. This represents one of the main impetuses to use postoperative ITF to quickly evaluate for a CSF leak when in question. Beta trace protein is a similar test with the advantage of providing a significantly faster result, but it is not available at our institution and has limited availability throughout the United States. 9 There is evidence that certain pneumocephalus patterns, locations, and volumes in postoperative imaging can aid in the diagnosis of leak presence; however, this method lacks a standardized protocol. 10 On the other hand, returning to the OR without a confirmatory test exposes patients to unnecessary risk and is an untenable approach from a cost perspective. There is a clear need for an effective and time-efficient modality to detect CSF leaks in the postoperative period. The use of ITF may be considered as a potential test to aid in this role.
ITF is widely used intraoperatively to localize CSF leaks and to confirm a watertight surgical closure, but its use in the postoperative setting as an indicator of CSF leak has not been described. Based on our cohort of 18 patients receiving 20 ITF injections in the postoperative period, we found this test to have 67% sensitivity, 100% specificity, 100% PPV, and 94% NPV. The low sensitivity reported here is likely due to the very low number of true CSF leaks in the ITF group (three). However, in a test such as ITF which is only performed when there is already significant clinical concern for the outcome of interest, the highest emphasis is placed on the specificity and NPV of the test, which were both found to be high. The specificity of ITF is comparable to what has been previously published for β-2 transferrin and β trace protein. In 2016, Oakley et al conducted a systematic review of diagnostic modalities for CSF rhinorrhea and found that across nine studies, β-2 transferrin had a sensitivity of 87 to 100% and a specificity of 71 to 100%, while across seven studies β trace protein had a sensitivity of 91 to 100% and a specificity of 86 to 100%. 9 ITF has the benefit over these modalities of providing rapid results without the need for specialized laboratory facilities. ITF also has a very low likelihood of producing false positive results (defined as the detection of fluorescein emanating from the surgical site in the absence of a true CSF leak) because CSF is the carrier of fluorescein and the dye itself is very distinct in appearance compared with biological tissues and products. This principle is demonstrated by the 100% specificity and PPV values calculated for this test. ITF injection does have the potential to infrequently produce false-negative results (defined as the failure to detect fluorescein emanating from the surgical site in the presence of a true CSF leak) as tissue edema, packing material, or poor patient tolerance of endoscopy may obscure fluorescein-stained CSF from view. There is also a potential for operator error if the fluorescein is not injected into the intrathecal space. Anatomic issues could arise as well, such as spinal stenosis, that could affect timely circulation of the fluorescein to the intracranial space. This is demonstrated by the comparatively lower sensitivity of ITF in this cohort.
One drawback of ITF as a diagnostic technique is the potential to expose patients to additional procedures (lumbar punctures / drains) which, despite the excellent safety profile of ITF itself, may expose patients to additional discomfort and risk. Lumbar puncture was not associated with postoperative meningitis and binary logistic regression did not demonstrate either ITF or placement of a postoperative lumbar drain to be independent risk factors for postoperative meningitis in this cohort, but the possibility for morbidity associated with these procedures must nevertheless be taken into consideration. Premedication with levetiracetam prior to ITF injection is also not completely devoid of risks, the most common of which are mild-to-moderate in severity, including asthenia and somnolence. 11 In the patients who received ITF, the risk of adverse effects was felt to be outweighed by the high likelihood that, in the absence of ITF, these patients would require a return trip to the OR to definitively rule out a CSF leak, or have a lumbar drain placed for 72 hours of CSF diversion. Compared with returning to the OR, ITF carries relatively low cost and risk.
This study has several limitations that warrant discussion. This investigation is limited by the small sample size of our cohort and the retrospective design of the project, which prevented a direct head-to-head comparison between ITF and other testing modalities. Bias was also introduced when calculating the PPV and NPVs for ITF because patients who underwent ITF injection were at increased clinical suspicion for CSF rhinorrhea. This may have improved the reported PPV and NPV of the test. To most accurately calculate PPV or NPV, all patients undergoing skull base surgery would need to receive ITF injection, which is unnecessary for the majority of patients. Additionally, ITF was not directly compared with a gold standard test but rather the patient's clinical course was ultimately the gold standard to determine the presence or absence of CSF leak, which potentially introduces bias. Finally, this study analyzed data gathered from a single rhinologist's cases, potentially limiting the generalizability of these findings as ITF requires bedside endoscopy, which is, to some degree, user-dependent. Future studies are needed to further elucidate ITF's place in the diagnostic algorithm for postoperative CSF leaks.
Conclusion
Diagnosing CSF leaks in the postoperative period following pituitary and ventral skull base surgery can be a challenging problem. Existing diagnostic modalities suffer from delayed result reporting, limited availability, or suboptimal sensitivity and specificity. The use of ITF may be considered as a potential test to aid in this role, and has demonstrated acceptable specificity and safety in this preliminary study.
Conflict of Interest None declared.
Note
Abstract was presented as a Virtual Poster, September 12 th 2020, American Rhinologic Society (Originally in Boston, MA).
References
- 1.Raza S M, Banu M A, Donaldson A, Patel K S, Anand V K, Schwartz T H. Sensitivity and specificity of intrathecal fluorescein and white light excitation for detecting intraoperative cerebrospinal fluid leak in endoscopic skull base surgery: a prospective study. J Neurosurg. 2016;124(03):621–626. doi: 10.3171/2014.12.JNS14995. [DOI] [PubMed] [Google Scholar]
- 2.Seth R, Rajasekaran K, Benninger M S, Batra P S. The utility of intrathecal fluorescein in cerebrospinal fluid leak repair. Otolaryngol Head Neck Surg. 2010;143(05):626–632. doi: 10.1016/j.otohns.2010.07.011. [DOI] [PubMed] [Google Scholar]
- 3.Jakimovski D, Bonci G, Attia M.Incidence and significance of intraoperative cerebrospinal fluid leak in endoscopic pituitary surgery using intrathecal fluorescein World Neurosurg 201482(3-4):e513–e523. [DOI] [PubMed] [Google Scholar]
- 4.Kirchner F R, Proud G O. Method for the identification and localization of cerebrospinal fluid, rhinorrhea and otorrhea. Laryngoscope. 1960;70:921–931. doi: 10.1288/00005537-196007000-00004. [DOI] [PubMed] [Google Scholar]
- 5.Keerl R, Weber R K, Draf W, Wienke A, Schaefer S D. Use of sodium fluorescein solution for detection of cerebrospinal fluid fistulas: an analysis of 420 administrations and reported complications in Europe and the United States. Laryngoscope. 2004;114(02):266–272. doi: 10.1097/00005537-200402000-00016. [DOI] [PubMed] [Google Scholar]
- 6.Flynn J P, Pavelonis A, Ledbetter L. The utility of computed tomography and intrathecal fluorescein in the management of cerebrospinal fluid leak. Am J Rhinol Allergy. 2020;34(03):342–347. doi: 10.1177/1945892419896243. [DOI] [PubMed] [Google Scholar]
- 7.Placantonakis D G, Tabaee A, Anand V K, Hiltzik D, Schwartz T H.Safety of low-dose intrathecal fluorescein in endoscopic cranial base surgeryNeurosurgery 2007;61(3, Suppl):161–165, discussion 165–166 [DOI] [PubMed]
- 8.Newcombe R G. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med. 1998;17(08):857–872. doi: 10.1002/(sici)1097-0258(19980430)17:8<857::aid-sim777>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
- 9.Oakley G M, Alt J A, Schlosser R J, Harvey R J, Orlandi R R. Diagnosis of cerebrospinal fluid rhinorrhea: an evidence-based review with recommendations. Int Forum Allergy Rhinol. 2016;6(01):8–16. doi: 10.1002/alr.21637. [DOI] [PubMed] [Google Scholar]
- 10.Banu M A, Szentirmai O, Mascarenhas L, Salek A A, Anand V K, Schwartz T H. Pneumocephalus patterns following endonasal endoscopic skull base surgery as predictors of postoperative CSF leaks. J Neurosurg. 2014;121(04):961–975. doi: 10.3171/2014.5.JNS132028. [DOI] [PubMed] [Google Scholar]
- 11.Lyseng-Williamson K A. Spotlight on levetiracetam in epilepsy. CNS Drugs. 2011;25(10):901–905. doi: 10.2165/11208340-000000000-00000. [DOI] [PubMed] [Google Scholar]