ORIGINAL ARTICLE
Perioperative continuous cerebrospinal fluid pressure monitoring in patients with spontaneous cerebrospinal fluid leaks Yanjun J. Xie, BA1 , Josef Shargorodsky, MD, MPH2 , Andrew P. Lane, MD2 , Masaru Ishii, MD, PhD2 , David Solomon, MD, PhD3 , Abhay Moghekar, MBBS3 , Gary L. Gallia, MD, PhD4 and Douglas D. Reh, MD2
Background: Elevated intracranial pressure (ICP) is an inciting factor for cerebrospinal fluid (CSF) leaks and can be measured by CSF pressure (CSFP) monitoring. Current CSFP literature is limited to the assessments of opening pressure. This study reinvestigates a previously discussed monitoring approach that evaluates continuous CSFP parameters, physiologic measurements, and treatment outcomes in patients undergoing endoscopic repair of spontaneous CSF leaks. Methods: Retrospective review of patients undergoing endoscopic endonasal repair of spontaneous CSF leaks. All participants had a lumbar catheter placed for 24-hour continuous preoperative pressure monitoring, and 24 hours of continuous monitoring starting 48 hours aer repair. In addition to patient characteristics, mean and peak CSFP, pulse waveform amplitudes (PWAs), and related parameters were calculated. Results: Twenty-five patients underwent monitoring between 2004 and 2013, with a mean follow-up of 526 days. The mean age was 49.2 years, the mean body mass index (BMI) 38.5, and 8 of 25 (32%) had obstructive sleep apnea. Although mean CSFP and PWA decreased aer the repair, mean peak CSFP increased by 1.56 cmH2 O (1.15 mmHg). Six
C
erebrospinal fluid (CSF) leaks result from anatomic disruption within the skull base and can present as watery rhinorrhea. CSF rhinorrhea is classified by the cause of the leak, such as tumor, trauma, or congeni-
1 Johns
Hopkins School of Medicine, Baltimore, MD; 2 Johns Hopkins Department of Otolaryngology–Head and Neck Surgery, Baltimore, MD; 3 Johns Hopkins Department of Neurology, Baltimore, MD; 4 Johns Hopkins Department of Neurosurgery, Baltimore, MD Correspondence to: Douglas D. Reh, MD, Department of Otolaryngology–Head and Neck Surgery, Johns Hopkins University School of Medicine, 601 North Caroline Street, 6th Floor, Baltimore, MD 21287-0910; e-mail: [email protected] Potential conflict of interest: None provided. Received: 22 April 2014; Revised: 14 August 2014; Accepted: 28 August 2014 DOI: 10.1002/alr.21424 View this article online at wileyonlinelibrary.com.
71
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
patients (24%) had elevation in their CSFP >25 cmH2 O (18.4 mmHg) for a minimum of 4% of the recording time. Based on their continuous pressure monitoring data, 9 patients (36%) underwent treatment for high ICP, either with acetazolamide or a ventricular shunt. There were no CSF leak recurrences. Conclusion: Continuous perioperative CSFP monitoring provides valuable insight into multiple physiologic parameters. Systematic continuous CSFP monitoring can identify individuals in need of ICP-lowering therapy, possibly imC 2014 proving the outcomes in CSF leak repair surgeries. ARS-AAOA, LLC.
Key Words: cerebrospinal fluid leak; cerebrospinal fluid rhinorrhea; idiopathic intracranial hypertension; intracranial pressure monitoring; endoscopic skull base surgery How to Cite this Article: Xie YJ, Shargorodsky J, Lane AP, et al. Perioperative continuous cerebrospinal fluid pressure monitoring in patients with spontaneous cerebrospinal fluid leaks. Int Forum Allergy Rhinol. 2015;5:71–77.
tal abnormalities.1 A CSF leak with unidentifiable etiology is referred to as a spontaneous CSF leak.2 Idiopathic intracranial hypertension (IIH) is increasingly recognized as a potential cause of spontaneous CSF leaks in the literature.3–7 Risk factors for IIH include the female gender, child-bearing and middle ages, and obesity. Patients often present with headaches, pulsatile tinnitus, or visual disturbances.3–5 Physical exam may reveal papilledema, indicating an elevated intracranial pressure (ICP). Highresolution computed tomography (CT) and magnetic resonance imaging (MRI) demonstrate distinguishing features, such as the presence of a skull-base defect, encephalocele, or empty sella.4 Several studies have also examined IIH as a cause of spontaneous CSF leaks through clinical evaluation using the modified Dandy criteria to stratify risk (Fig. 1).8
Xie et al.
FIGURE 1. The modified Dandy criteria for diagnosing IIH.8 CN VI = cranial nerve 6; CSF = cerebrospinal fluid; CT = computed tomography; ICP = intracranial pressure; IIH = intracranial hypertension.
Only a few studies have measured CSF pressure (CSFP) directly, either with opening pressure at the time of lumbar drain placement or with perioperative monitoring.9, 10 These studies have been limited in that they use single or spot measurements that are static assessments of CSFP and do not account for positional, diurnal, or physiologic variations of CSFP. The isolated measurements can therefore yield falsely elevated pressures.11–13 Despite the limitations, studies show that patients with spontaneous CSF leaks have persistently elevated ICP after successful surgical repair.9, 10 Spontaneous CSF leak patients also have the highest recurrence rate (25% to 87%), compared to 25 cmH2 O or 18.4 mmHg; Fig. 3). Percent of CSFP above the modified Dandy criteria was analyzed for each set of monitoring data. CSFP was recorded in millimeters of mercury (mmHg) and reported in mmHg
Xie et al.
FIGURE 3. AUC 25 (shaded area) represents an integrated measurement of both mean CSFP and time for CSFP above the modified Dandy criteria (CSFP > 18.4 mmHg, or 25 cmH2 O). AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; ICP = intracranial pressure; OSA = obstructive sleep apnea.
TABLE 1. Patient demographic and clinical data∗
and centimeters of water (cmH2 O) for the analysis in this article (cmH2 O = mmHg × 1.36). After data collection, a descriptive analysis of patient characteristics was performed with calculation of means for continuous variables and proportions for categorical variables. Changes between preoperative and postoperative CSFP parameters were examined using the 2-tailed paired t test. CSFP differences between patients who received postoperative ICP-lowering therapy and those who did not were analyzed using the 2-tailed independent t test with STATA 12.0 (StataCorp, College Station, TX). A p < 0.05 was considered statistically significant for all analyses. Treatment outcomes after the endoscopic repair were retrospectively reviewed including complications, additional ICP-lowering medical or surgical management, and leak recurrence. After surgery, patients were followed at the Johns Hopkins Otolaryngology–Head and Neck Surgery or the Neurology and Neurosurgery outpatient clinics.
n
%
Female
24
96
Male
1
4
Demographics
Age (years), mean ± SD
49.2 ± 12.5
BMI (kg/m ), mean ± SD
38.5 ± 11.7
2
Clinical information Headaches
12
48
OSA
8
32
Empty sella
8
44a
Pulsatile tinnitus
4
16
Preoperative papilledema
4
16
Cribriform
9
36
Sphenoid sinus
9
36
Ethmoid sinus
4
16
Frontal sinus
2
8
Multiple locations
1
4
Temporalis fascia
14
56
Mucosal graft
6
24
Septal flap
2
8
Multiple types
2
8
Synthetic materials
1
4
Surgical findings: defect location
Results Twenty-five patients (24 females and 1 male) with spontaneous CSF rhinorrhea underwent perioperative monitoring between 2004 and 2013. Patient demographics and clinical data are detailed in Table 1. The mean ± standard deviation (SD) age was 49.2 ± 12.5 years, and the mean BMI was 38.5 ± 11.7. The prevalence of signs and symptoms related to elevated ICP was assessed: 12 (48%) had preoperative headaches; 8 (32%) had OSA; 4 (16%) had pulsatile tinnitus; and 4 (16%) had papilledema on ophthalmologic exam. Empty sella was identified on high-resolution MRI by both radiologists and the surgical team and was present in 8 of 18 subjects (44%) with available MRI scans. All patients had an encephalocele as well as an active CSF leak at the time of repair. The most common leak locations were the cribriform plate (36%) and sphenoid sinus (36%). One of the 25 patients had CSF leaks in multiple locations (cribriform plate and sphenoid roof).
Surgical findings: repair materials
∗
Values are n (%) unless otherwise indicated. Eighteen of 25 patients in this study had available MRI scans. Empty sella was present in 8 of 18 (44%) patients. BMI = body mass index; OSA = obstructive sleep apnea; SD = standard deviation. a
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
74
Continuous CSF pressure monitoring
TABLE 2. Preoperative and postoperative CSFP monitoring data Preoperative
Postoperative
Average of preoperative/postoperative differences
p
Mean CSFP (cmH2 O/mmHg)
8.06/5.93
7.36/5.41
−1.17/−0.86
0.531
Peak CSFP (cmH2 O/mmHg)
37.8/27.8
39.8/29.3
1.56/1.15
0.798
PWA (cmH2 O/mmHg)
6.01/4.42
5.57/4.10
−0.83/−0.61
0.183
1.25
3.19
2.36
0.098
21.8/16.0
142.2/104.6
165.2/121.5
0.079
% of Data above the modified Dandy criteria (>25cmH2 O) AUC 25 (cmH2 O-min/mmHg-min)
AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; PWA = pulse waveform amplitude.
Table 2 displays the preoperative and postoperative CSFP monitoring data. All postoperative monitoring data were recorded prior to additional ICP-lowering therapies. The preoperative mean CSFP was 8.06 cmH2 O (5.93 mmHg) and the postoperative mean was 7.36 cmH2 O (5.41 mmHg). The average change in mean CSFP was a decrease of 1.17 cmH2 O (0.86 mmHg). The preoperative pulse waveform amplitude (PWA) was 6.01 cmH2 O (4.42 mmHg) and the postoperative PWA was 5.57 cmH2 O (4.10 mmHg). Although mean CSFP and PWA decreased after the repair, the average change in peak CSFP increased from 37.8 cmH2 O (27.8 mmHg) to 39.8 cmH2 O (29.3 mmHg). The percentage of CSFP that exceeded the modified Dandy criteria was calculated. Preoperatively, 1.25% of CSFP during a continuous monitoring was above these criteria. This increased to 3.19% postoperatively. The p values for these changes are included in Table 2. Six subjects (24%) had elevation in their CSFP >25 cmH2 O (18.4 mmHg) for a minimum of 4% of the recording time. Postoperatively, 4 (16%) patients had papilledema. One subject developed new-onset papilledema, whereas the remaining 3 had persisting symptoms after surgery. Eight patients (32%) received acetazolamide, and 2 (8%) underwent shunt placement to lower ICP. Table 3 summarizes the CSFP parameters for each treatment group. Nine patients (36%) in total required postoperative ICP-lowering therapy, 1 of which received both diuretic and surgical management. One patient in the study deviated from the protocol and underwent shunt placement without acetazolamide therapy. After the repair surgery, this patient developed multiple episodes of unstable and abnormally elevated CSFP with a peak CSFP of 54.1 cmH2 O (39.8 mmHg) and wide pulse amplitudes. She also complained of significant postoperative headaches. Given these findings, our neurology team believed that the patient might be a poor candidate for acetazolamide therapy. Upon discussion with the medical team, the patient elected to directly undergo ventricular shunt placement instead. CSFP parameters were compared between the group of patients who received postoperative treatment and the group that did not. In general, these parameters were higher in the cohort of patients who received additional ICP therapies. Mean CSFP was 4.81 cmH2 O (3.54 mmHg) in the no-treatment group and 11.3 cmH2 O (8.31 mmHg) in the
75
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
treatment group (p = 0.095). Peak CSFP was 33.1 cmH2 O (24.3 mmHg) in the no-treatment group and 50.3 cmH2 O (37.0 mmHg) in the treatment group (p = 0.070). Likewise, PWA increased from 4.99 cmH2 O (3.67 mmHg) to 6.47 cmH2 O (4.76 mmHg) (p = 0.120), and the percent of CSFP data above the modified Dandy criteria increased from 1.27% to 6.19% (p = 0.058). The postoperative AUC 25 was 94.2 cmH2 O-min (69.3 mmHg-min) in the notreatment cohort and 174.1 cmH2 O-min (128 mmHg-min) in the treatment group (p = 0.540). The average follow-up was 17 months (range, 1 to 65 months), with 9 subjects being followed up in 2013. Patients who received additional treatments were followed for an average of 423 days, and those who did not were followed for an average of 584 days. Seven (43.8%) of 16 patients who did not receive additional therapies had follow-up longer than 1 year. During the study period, no patients developed a recurrent CSF leak or had a secondary leak at other sites of the skull base.
Discussion Elevated ICP is increasingly recognized as an inciting factor for patients with spontaneous CSF leaks.1–7 The endoscopic approach for repairing spontaneous CSF leaks has become the standard of care.15 Despite improvement in surgical techniques, patients with spontaneous CSF leaks present numerous challenges. Spontaneous CSF leak patients often have persistent ICP elevation after successful repair. They also have the highest recurrence rate compared to other etiologies.3, 4, 10 Perioperative CSFP monitoring is helpful for guiding treatment options in this group of patients. Current literature utilizing measurements of CSFP is limited to the assessment of opening pressures. These measurements are static and do not account for variations of CSFP overnight when more sustained and pathologic elevations of CSFP are seen. This study reinvestigates our previously discussed monitoring approach that evaluated continuous CSFP parameters, adding the analyses of physiologic measurements and treatment outcomes in patients undergoing endoscopic repair of spontaneous CSF leaks. Demographic features and clinical presentations of participants in this study are consistent with previous spontaneous CSF leaks literature. In a review of 23 articles
Xie et al.
TABLE 3. Postoperative CSFP parameters correlated with ICP-lowering treatments No additional treatment
Additional ICP-lowering treatmenta
Difference
p
16 (64)
9 (36)
7 (28)
n/a
Mean CSFP (cmH2 O/mmHg)
4.81/3.54
11.3/8.31
6.49/4.77
0.095
Peak CSFP (cmH2 O/mmHg)
33.1/24.3
50.3/37.0
17.2/12.6
0.070
PWA (cmH2 O/mmHg)
4.99/3.67
6.47/4.76
1.48/1.09
0.120
1.27
6.19
4.92
0.058
94.2/69.3
174.1/128
79.9/58.7
0.540
Patients, n (%)
% of Data above the modified Dandy criteria AUC 25 (cmH2 O-min/mmHg-min) a
This cohort includes both acetazolamide therapy and ventricular shunting. AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; ICP = intracranial pressure; PWA = pulse waveform amplitude.
by Perez et al.,7 characteristics most commonly associated with IIH were as follows: females of child-bearing or middle ages; high BMI; and a constellation of signs and symptoms including CSF rhinorrhea, headaches, tinnitus, visual disturbances, papilledema, and the presence of empty sella on imaging. On imaging, all patients had meningoencephalocele herniating through a skull-base defect, and 44% had empty sella. The majority of patients were middle-aged female and obese with a mean BMI of 38.5. Major signs and symptoms were the following: CSF rhinorrhea (100%), headaches (48%), past medical history of OSA (32%), pulsatile tinnitus (16%), and papilledema (16%). One of 4 patients in the study developed new-onset papilledema after surgery, whereas the remaining 3 had existing preoperative papilledema. This finding could be attributed to persistent elevation of ICP after the repair, manifested as unresolved papilledema. Because of its gradual onset, papilledema has relatively poor diagnostic value for predicting acute intracranial hypertension.16 These findings do, however, support careful postoperative ophthalmologic evaluation following CSF leak repair. OSA was another important risk factor associated with intracranial hypertension. Although 32% of our patients had OSA per medical records, the severity of OSA was not elicited or confirmed by polysomnography as part of the study protocol. Future investigation will be directed toward elucidating any correlation between OSA severity and CSFP changes in this patient population. Continuous CSFP monitoring has the potential to provide useful information for managing patients undergoing the repair of spontaneous CSF leaks. Prior studies assessed CSFP at single time points, which did not take into account ICP changes due to patient positioning or physiologic CSFP changes overnight.11, 12 The continuous monitoring approach is unique in that it enables ongoing detection of abnormal pressure elevations and correlates these changes with physiologic events in real-time. Figure 2 displays a sample recording for a patient with OSA. As the respiratory rate decreased around 12:13 AM (onset of an apnea episode), the ICP increased accordingly followed by a decline in oxygen saturation from 100% to approximately 75%. After the patient regained normal breathing pattern,
ICP lowered to baseline with a return of oxygen saturation. Simultaneous capturing of CSFP and other physiologic measurements offers a dynamic representation of changes in ICP over the patient’s postoperative course. This provides our neurologists with useful clinical information to determine whether additional treatment is required to manage a patient’s ICP after surgical repair. The CSFP measured in this study were lower than those reported in other studies. Unlike that in Illing et al.17 (mean ICP: 27.7 cmH2 O) and Schlosser et al.10 (mean ICP: 32.5 cmH2 O), the pressures in our study were mostly less than 25 cmH2 O. As mentioned previously, opening pressures can be artificially elevated due to concurrent Valsalva maneuvers or other physiologic changes at the time of lumbar drain placement. For this reason, opening pressures were not routinely recorded in our study. Although many patients actually had high postoperative CSFP in the initial recording period, their pressures gradually leveled off throughout the night. Interestingly, another study that used continuous CSFP monitoring similar to our technique also reported a mean postoperative CSFP less than 25 cmH2 O (mean ICP: 15 cmH2 O).18 In addition, continuous CSFP monitoring provides a way to compare preoperative and postoperative CSFP parameters: mean CSFP, peak CSFP, PWA, and AUC 25. In our study, most CSFP parameters increased, whereas mean CSFP and PWA decreased slightly (mean CSFP change: −1.17 cmH2 O or −0.86 mmHg, p = 0.531; peak CSFP change: 1.56 cmH2 O or 1.15 mmHg, p = 0.798; PWA change: −0.83 cmH2 O or −0.61 mmHg, p = 0.183). However, the preoperative and postoperative changes were statistically insignificant. We could not conclude a physiologic increase or decrease of CSFP parameters based on these differences. Rather, they might reflect that CSFP did not alter significantly with the repair or at least directly after the repair. Alternatively, there may not have been enough time for sufficient CSFP elevation to develop after the immediate postoperative period when the pressure measurements were taken. These findings were consistent with studies that showed persistent ICP elevation despite successful repair.9, 10 Although mean CSFP, peak CSFP, and PWA provided useful information on the patient’s ICP
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
76
Continuous CSF pressure monitoring
status, they were single-dimensional pressure variables that did not factor in changes over time. AUC 25 was a timepressure product of CSFP above a certain threshold and is theoretically a more robust way to represent ICP changes. Although the change in AUC 25 was statistically insignificant (p = 0.079), its value increased considerably from 21.8 cmH2 O-min (16.0 mmHg-min) before the repair to 142.2 cmH2 O-min (104.6 mmHg-min) after the repair. Future studies will further assess the validity of AUC measurements in evaluating ICP status. In our practice, continuous CSFP monitoring is used to determine the need for postoperative ICP-lowering therapy to prevent CSF leak recurrences. Acetazolamide decreases CSF production at the choroid plexus and has been effective in reducing ICP via this mechanism.19 Acetazolamide therapy was started only after considering both the patient’s clinical status (presence of signs and symptoms consistent with elevated ICP) and the quantitative CSFP data. In the present study, 8 of 25 (32%) received acetazolamide after the surgery, and 2 patients (8%) underwent ventricular shunt placement (1 had both acetazolamide and a subsequent shunt). Although about one-third of patients required further ICP-lowering therapy, the majority of patients in the study did not need additional interventions. Therefore, prophylactic acetazolamide therapy was not initiated, considering the risks and benefits of starting every patient on this medication. CSFP parameters including mean CSFP, peak CSFP, PWA, and percentage of data above the Dandy criteria were overall higher in patients who received acetazolamide therapy and shunts compared to those who had no further treatment. This was expected given that patients receiving additional ICP-lowering therapies typically showed persistent ICP elevation after their surgery. Of note, higher doses of acetazolamide (1 to 2 g/day) were used for treatment compared to the commonly described dose. If the patient failed to respond to medical therapy, ventricular shunt was considered to further reduce ICP. All study patients were followed postoperatively in our outpa-
tient clinics with an average follow-up time of 526 days. Despite the high recurrence rate reported in the literature, no patients in this study developed a recurrent leak during the study period.3, 4, 10 The strengths and limitations of this study should be considered. The current study expanded from our previous cohort of 12 patients to include 25 who were treated in the same tertiary academic center.14 Although this was one of the larger series in the literature, the small number of participants allowed for only limited power in statistical analyses. All of the operations were performed at a single center, which had the advantage of maintaining consistency among surgical and medical approaches but also limited generalizability. In addition, although the monitoring was approached prospectively, the retrospective nature of data collection in this study only allowed for analysis of data that were already available. Due to a lack of comparison groups, it would be difficult to assess whether the high success rate in our patients was due to CSFP-lowering therapeutic interventions, as determined by the monitoring parameters, or to surgical techniques.
Conclusion In summary, continuous CSFP monitoring may provide clinically important information for managing patients who undergo endoscopic CSF leak repair surgery. It enables providers to correlate physiologic events with otherwise undetected changes in ICP over time and may aid in clinical decision-making for prescribing additional ICP-lowering therapies. Although prospective data with a control group is needed, this study nevertheless demonstrates the potential usefulness of the continuous CSFP monitoring approach. Identification of individuals with persistent ICP elevations after surgery through perioperative pressure monitoring may help in improving patient outcomes and surgical success in patients undergoing spontaneous anterior skull-base CSF leak repair.
References 1. 2.
3.
4.
5.
6.
7.
77
Ommaya AK. Cerebrospinal fluid rhinorrhea. Neurology. 1964;14:106–113. O’Connell JE. Primary spontaneous cerebrospinal fluid rhinorrhea. J Neurol Nerosurg Psychiatry. 1964;27:241–246. Yang Z, Wang B, Wang C, Liu P. Primary spontaneous cerebrospinal fluid rhinorrhea: a symptom of idiopathic intracranial hypertension? J Neurosurg. 2011;115:165–170. Chaaban MR, Illing E, Riley KO, Woodworth BA. Spontaneous cerebrospinal fluid leak repair: a five-year prospective evaluation. Laryngoscope. 2014;124:70–75. Brainard L, Chen DA, Aziz KM, Hillman TA. Association of benign intracranial hypertension and spontaneous encephalocele with cerebrospinal fluid leak. Otol Neurotol. 2012;33:1621–1624. Suryadevara AC, Fattal M, Woods CI. Nontraumatic cerebrospinal fluid rhinorrhea as a result of pseudotumor cerebri. Am J Otolaryngol. 2007;28:242–246. Perez MA, Bialer OY, Bruce BB, Newman NJ, Biousse V. Primary spontaneous cerebrospinal fluid leaks and idiopathic intracranial hypertension. J Neuroophthalmol. 2013;33:330–337.
8.
9.
10.
11.
12.
13.
14.
Shaw GY, Million SK. Benign intracranial hypertension: a diagnostic dilemma. Case Rep Otolaryngol. 2012;2012:814696. Schlosser RJ, Wilensky EM, Grady MS, Bolger WE. Elevated intracranial pressures in spontaneous cerebrospinal fluid leaks. Am J Rhinol. 2003;17:191–195. Schlosser RJ, Wilensky EM, Grady MS, Palmer JN, Kennedy DW, Bolger WE. Cerebrospinal fluid pressure monitoring after repair of cerebrospinal fluid leaks. Otolaryngol Head Neck Surg. 2004;130:443– 448. Poca MA, Sahuquillo J, Topczewski T, Lastra R, Font ML, Corral E. Posture-induced changes in intracranial pressure: a comparative study in patients with and without fluid blocks at the craniovertebral junction. Neurosurgery. 2006;58:899–906. Kropyvnytskyy IV, Saunders FW, Klemfuss H. Circadian rhythm of cerebral perfusion pressure and intracranial pressure. Brain Inj. 1999;13:45–52. Ng I, Lim J, Wong HB. Effects of head posture on cerebral hemodynamics: its influences on intracranial pressure, cerebral perfusion pressure, and cerebral oxygen. Neurosurgery. 2004;54:593–597, discussion 598. Reh DD, Gallia GL, Ramanathan M, et al. Perioperative continuous cerebrospinal fluid pressure monitor-
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
15.
16.
17.
18.
19.
ing in patients with spontaneous cerebrospinal fluid leaks: Presentation of a novel technique. Am J Rhinol Allergy. 2010;24:238–243. Hegazy HM, Carrau RL, Synderman, CH, Kassam A, Zweig J. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope. 2000;110:1166–1172. Steffen H, Eifert B, Aschoff A, Kolling GH, Volcker ¨ HE. The diagnostic value of optic disc evaluation in acute elevated intracranial pressure. Ophthalmology. 1996;103:1229–1232. Illing E, Schlosser RJ, Palmer JN, Cur´e J, Fox N, Woodworth BA. Spontaneous sphenoid lateral recess cerebrospinal fluid leaks arise from intracranial hypertension, not Sternberg’s canal. Int Forum Allergy Rhinol. 2014;4:246–250. Ramakrishnan VR, Suh JD, Chiu AG, Palmer JN. Reliability of preoperative assessment of cerebrospinal fluid pressure in the management of spontaneous cerebrospinal fluid leaks and encephaloceles. Int Forum Allergy Rhinol. 2011;1:201–205. Lee AG, Wall M. Papilledema: are we any nearer to a consensus on pathogenesis and treatment? Curr Neurol Neurosci Rep. 2012;12:334–339.
Perioperative continuous cerebrospinal fluid pressure monitoring in patients with spontaneous cerebrospinal fluid leaks Yanjun J. Xie, BA1 , Josef Shargorodsky, MD, MPH2 , Andrew P. Lane, MD2 , Masaru Ishii, MD, PhD2 , David Solomon, MD, PhD3 , Abhay Moghekar, MBBS3 , Gary L. Gallia, MD, PhD4 and Douglas D. Reh, MD2
Background: Elevated intracranial pressure (ICP) is an inciting factor for cerebrospinal fluid (CSF) leaks and can be measured by CSF pressure (CSFP) monitoring. Current CSFP literature is limited to the assessments of opening pressure. This study reinvestigates a previously discussed monitoring approach that evaluates continuous CSFP parameters, physiologic measurements, and treatment outcomes in patients undergoing endoscopic repair of spontaneous CSF leaks. Methods: Retrospective review of patients undergoing endoscopic endonasal repair of spontaneous CSF leaks. All participants had a lumbar catheter placed for 24-hour continuous preoperative pressure monitoring, and 24 hours of continuous monitoring starting 48 hours aer repair. In addition to patient characteristics, mean and peak CSFP, pulse waveform amplitudes (PWAs), and related parameters were calculated. Results: Twenty-five patients underwent monitoring between 2004 and 2013, with a mean follow-up of 526 days. The mean age was 49.2 years, the mean body mass index (BMI) 38.5, and 8 of 25 (32%) had obstructive sleep apnea. Although mean CSFP and PWA decreased aer the repair, mean peak CSFP increased by 1.56 cmH2 O (1.15 mmHg). Six
C
erebrospinal fluid (CSF) leaks result from anatomic disruption within the skull base and can present as watery rhinorrhea. CSF rhinorrhea is classified by the cause of the leak, such as tumor, trauma, or congeni-
1 Johns
Hopkins School of Medicine, Baltimore, MD; 2 Johns Hopkins Department of Otolaryngology–Head and Neck Surgery, Baltimore, MD; 3 Johns Hopkins Department of Neurology, Baltimore, MD; 4 Johns Hopkins Department of Neurosurgery, Baltimore, MD Correspondence to: Douglas D. Reh, MD, Department of Otolaryngology–Head and Neck Surgery, Johns Hopkins University School of Medicine, 601 North Caroline Street, 6th Floor, Baltimore, MD 21287-0910; e-mail: [email protected] Potential conflict of interest: None provided. Received: 22 April 2014; Revised: 14 August 2014; Accepted: 28 August 2014 DOI: 10.1002/alr.21424 View this article online at wileyonlinelibrary.com.
71
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
patients (24%) had elevation in their CSFP >25 cmH2 O (18.4 mmHg) for a minimum of 4% of the recording time. Based on their continuous pressure monitoring data, 9 patients (36%) underwent treatment for high ICP, either with acetazolamide or a ventricular shunt. There were no CSF leak recurrences. Conclusion: Continuous perioperative CSFP monitoring provides valuable insight into multiple physiologic parameters. Systematic continuous CSFP monitoring can identify individuals in need of ICP-lowering therapy, possibly imC 2014 proving the outcomes in CSF leak repair surgeries. ARS-AAOA, LLC.
Key Words: cerebrospinal fluid leak; cerebrospinal fluid rhinorrhea; idiopathic intracranial hypertension; intracranial pressure monitoring; endoscopic skull base surgery How to Cite this Article: Xie YJ, Shargorodsky J, Lane AP, et al. Perioperative continuous cerebrospinal fluid pressure monitoring in patients with spontaneous cerebrospinal fluid leaks. Int Forum Allergy Rhinol. 2015;5:71–77.
tal abnormalities.1 A CSF leak with unidentifiable etiology is referred to as a spontaneous CSF leak.2 Idiopathic intracranial hypertension (IIH) is increasingly recognized as a potential cause of spontaneous CSF leaks in the literature.3–7 Risk factors for IIH include the female gender, child-bearing and middle ages, and obesity. Patients often present with headaches, pulsatile tinnitus, or visual disturbances.3–5 Physical exam may reveal papilledema, indicating an elevated intracranial pressure (ICP). Highresolution computed tomography (CT) and magnetic resonance imaging (MRI) demonstrate distinguishing features, such as the presence of a skull-base defect, encephalocele, or empty sella.4 Several studies have also examined IIH as a cause of spontaneous CSF leaks through clinical evaluation using the modified Dandy criteria to stratify risk (Fig. 1).8
Xie et al.
FIGURE 1. The modified Dandy criteria for diagnosing IIH.8 CN VI = cranial nerve 6; CSF = cerebrospinal fluid; CT = computed tomography; ICP = intracranial pressure; IIH = intracranial hypertension.
Only a few studies have measured CSF pressure (CSFP) directly, either with opening pressure at the time of lumbar drain placement or with perioperative monitoring.9, 10 These studies have been limited in that they use single or spot measurements that are static assessments of CSFP and do not account for positional, diurnal, or physiologic variations of CSFP. The isolated measurements can therefore yield falsely elevated pressures.11–13 Despite the limitations, studies show that patients with spontaneous CSF leaks have persistently elevated ICP after successful surgical repair.9, 10 Spontaneous CSF leak patients also have the highest recurrence rate (25% to 87%), compared to 25 cmH2 O or 18.4 mmHg; Fig. 3). Percent of CSFP above the modified Dandy criteria was analyzed for each set of monitoring data. CSFP was recorded in millimeters of mercury (mmHg) and reported in mmHg
Xie et al.
FIGURE 3. AUC 25 (shaded area) represents an integrated measurement of both mean CSFP and time for CSFP above the modified Dandy criteria (CSFP > 18.4 mmHg, or 25 cmH2 O). AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; ICP = intracranial pressure; OSA = obstructive sleep apnea.
TABLE 1. Patient demographic and clinical data∗
and centimeters of water (cmH2 O) for the analysis in this article (cmH2 O = mmHg × 1.36). After data collection, a descriptive analysis of patient characteristics was performed with calculation of means for continuous variables and proportions for categorical variables. Changes between preoperative and postoperative CSFP parameters were examined using the 2-tailed paired t test. CSFP differences between patients who received postoperative ICP-lowering therapy and those who did not were analyzed using the 2-tailed independent t test with STATA 12.0 (StataCorp, College Station, TX). A p < 0.05 was considered statistically significant for all analyses. Treatment outcomes after the endoscopic repair were retrospectively reviewed including complications, additional ICP-lowering medical or surgical management, and leak recurrence. After surgery, patients were followed at the Johns Hopkins Otolaryngology–Head and Neck Surgery or the Neurology and Neurosurgery outpatient clinics.
n
%
Female
24
96
Male
1
4
Demographics
Age (years), mean ± SD
49.2 ± 12.5
BMI (kg/m ), mean ± SD
38.5 ± 11.7
2
Clinical information Headaches
12
48
OSA
8
32
Empty sella
8
44a
Pulsatile tinnitus
4
16
Preoperative papilledema
4
16
Cribriform
9
36
Sphenoid sinus
9
36
Ethmoid sinus
4
16
Frontal sinus
2
8
Multiple locations
1
4
Temporalis fascia
14
56
Mucosal graft
6
24
Septal flap
2
8
Multiple types
2
8
Synthetic materials
1
4
Surgical findings: defect location
Results Twenty-five patients (24 females and 1 male) with spontaneous CSF rhinorrhea underwent perioperative monitoring between 2004 and 2013. Patient demographics and clinical data are detailed in Table 1. The mean ± standard deviation (SD) age was 49.2 ± 12.5 years, and the mean BMI was 38.5 ± 11.7. The prevalence of signs and symptoms related to elevated ICP was assessed: 12 (48%) had preoperative headaches; 8 (32%) had OSA; 4 (16%) had pulsatile tinnitus; and 4 (16%) had papilledema on ophthalmologic exam. Empty sella was identified on high-resolution MRI by both radiologists and the surgical team and was present in 8 of 18 subjects (44%) with available MRI scans. All patients had an encephalocele as well as an active CSF leak at the time of repair. The most common leak locations were the cribriform plate (36%) and sphenoid sinus (36%). One of the 25 patients had CSF leaks in multiple locations (cribriform plate and sphenoid roof).
Surgical findings: repair materials
∗
Values are n (%) unless otherwise indicated. Eighteen of 25 patients in this study had available MRI scans. Empty sella was present in 8 of 18 (44%) patients. BMI = body mass index; OSA = obstructive sleep apnea; SD = standard deviation. a
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
74
Continuous CSF pressure monitoring
TABLE 2. Preoperative and postoperative CSFP monitoring data Preoperative
Postoperative
Average of preoperative/postoperative differences
p
Mean CSFP (cmH2 O/mmHg)
8.06/5.93
7.36/5.41
−1.17/−0.86
0.531
Peak CSFP (cmH2 O/mmHg)
37.8/27.8
39.8/29.3
1.56/1.15
0.798
PWA (cmH2 O/mmHg)
6.01/4.42
5.57/4.10
−0.83/−0.61
0.183
1.25
3.19
2.36
0.098
21.8/16.0
142.2/104.6
165.2/121.5
0.079
% of Data above the modified Dandy criteria (>25cmH2 O) AUC 25 (cmH2 O-min/mmHg-min)
AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; PWA = pulse waveform amplitude.
Table 2 displays the preoperative and postoperative CSFP monitoring data. All postoperative monitoring data were recorded prior to additional ICP-lowering therapies. The preoperative mean CSFP was 8.06 cmH2 O (5.93 mmHg) and the postoperative mean was 7.36 cmH2 O (5.41 mmHg). The average change in mean CSFP was a decrease of 1.17 cmH2 O (0.86 mmHg). The preoperative pulse waveform amplitude (PWA) was 6.01 cmH2 O (4.42 mmHg) and the postoperative PWA was 5.57 cmH2 O (4.10 mmHg). Although mean CSFP and PWA decreased after the repair, the average change in peak CSFP increased from 37.8 cmH2 O (27.8 mmHg) to 39.8 cmH2 O (29.3 mmHg). The percentage of CSFP that exceeded the modified Dandy criteria was calculated. Preoperatively, 1.25% of CSFP during a continuous monitoring was above these criteria. This increased to 3.19% postoperatively. The p values for these changes are included in Table 2. Six subjects (24%) had elevation in their CSFP >25 cmH2 O (18.4 mmHg) for a minimum of 4% of the recording time. Postoperatively, 4 (16%) patients had papilledema. One subject developed new-onset papilledema, whereas the remaining 3 had persisting symptoms after surgery. Eight patients (32%) received acetazolamide, and 2 (8%) underwent shunt placement to lower ICP. Table 3 summarizes the CSFP parameters for each treatment group. Nine patients (36%) in total required postoperative ICP-lowering therapy, 1 of which received both diuretic and surgical management. One patient in the study deviated from the protocol and underwent shunt placement without acetazolamide therapy. After the repair surgery, this patient developed multiple episodes of unstable and abnormally elevated CSFP with a peak CSFP of 54.1 cmH2 O (39.8 mmHg) and wide pulse amplitudes. She also complained of significant postoperative headaches. Given these findings, our neurology team believed that the patient might be a poor candidate for acetazolamide therapy. Upon discussion with the medical team, the patient elected to directly undergo ventricular shunt placement instead. CSFP parameters were compared between the group of patients who received postoperative treatment and the group that did not. In general, these parameters were higher in the cohort of patients who received additional ICP therapies. Mean CSFP was 4.81 cmH2 O (3.54 mmHg) in the no-treatment group and 11.3 cmH2 O (8.31 mmHg) in the
75
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
treatment group (p = 0.095). Peak CSFP was 33.1 cmH2 O (24.3 mmHg) in the no-treatment group and 50.3 cmH2 O (37.0 mmHg) in the treatment group (p = 0.070). Likewise, PWA increased from 4.99 cmH2 O (3.67 mmHg) to 6.47 cmH2 O (4.76 mmHg) (p = 0.120), and the percent of CSFP data above the modified Dandy criteria increased from 1.27% to 6.19% (p = 0.058). The postoperative AUC 25 was 94.2 cmH2 O-min (69.3 mmHg-min) in the notreatment cohort and 174.1 cmH2 O-min (128 mmHg-min) in the treatment group (p = 0.540). The average follow-up was 17 months (range, 1 to 65 months), with 9 subjects being followed up in 2013. Patients who received additional treatments were followed for an average of 423 days, and those who did not were followed for an average of 584 days. Seven (43.8%) of 16 patients who did not receive additional therapies had follow-up longer than 1 year. During the study period, no patients developed a recurrent CSF leak or had a secondary leak at other sites of the skull base.
Discussion Elevated ICP is increasingly recognized as an inciting factor for patients with spontaneous CSF leaks.1–7 The endoscopic approach for repairing spontaneous CSF leaks has become the standard of care.15 Despite improvement in surgical techniques, patients with spontaneous CSF leaks present numerous challenges. Spontaneous CSF leak patients often have persistent ICP elevation after successful repair. They also have the highest recurrence rate compared to other etiologies.3, 4, 10 Perioperative CSFP monitoring is helpful for guiding treatment options in this group of patients. Current literature utilizing measurements of CSFP is limited to the assessment of opening pressures. These measurements are static and do not account for variations of CSFP overnight when more sustained and pathologic elevations of CSFP are seen. This study reinvestigates our previously discussed monitoring approach that evaluated continuous CSFP parameters, adding the analyses of physiologic measurements and treatment outcomes in patients undergoing endoscopic repair of spontaneous CSF leaks. Demographic features and clinical presentations of participants in this study are consistent with previous spontaneous CSF leaks literature. In a review of 23 articles
Xie et al.
TABLE 3. Postoperative CSFP parameters correlated with ICP-lowering treatments No additional treatment
Additional ICP-lowering treatmenta
Difference
p
16 (64)
9 (36)
7 (28)
n/a
Mean CSFP (cmH2 O/mmHg)
4.81/3.54
11.3/8.31
6.49/4.77
0.095
Peak CSFP (cmH2 O/mmHg)
33.1/24.3
50.3/37.0
17.2/12.6
0.070
PWA (cmH2 O/mmHg)
4.99/3.67
6.47/4.76
1.48/1.09
0.120
1.27
6.19
4.92
0.058
94.2/69.3
174.1/128
79.9/58.7
0.540
Patients, n (%)
% of Data above the modified Dandy criteria AUC 25 (cmH2 O-min/mmHg-min) a
This cohort includes both acetazolamide therapy and ventricular shunting. AUC 25 = area under the curve for CSFP > 25cmH2 O; CSFP = cerebrospinal fluid pressure; ICP = intracranial pressure; PWA = pulse waveform amplitude.
by Perez et al.,7 characteristics most commonly associated with IIH were as follows: females of child-bearing or middle ages; high BMI; and a constellation of signs and symptoms including CSF rhinorrhea, headaches, tinnitus, visual disturbances, papilledema, and the presence of empty sella on imaging. On imaging, all patients had meningoencephalocele herniating through a skull-base defect, and 44% had empty sella. The majority of patients were middle-aged female and obese with a mean BMI of 38.5. Major signs and symptoms were the following: CSF rhinorrhea (100%), headaches (48%), past medical history of OSA (32%), pulsatile tinnitus (16%), and papilledema (16%). One of 4 patients in the study developed new-onset papilledema after surgery, whereas the remaining 3 had existing preoperative papilledema. This finding could be attributed to persistent elevation of ICP after the repair, manifested as unresolved papilledema. Because of its gradual onset, papilledema has relatively poor diagnostic value for predicting acute intracranial hypertension.16 These findings do, however, support careful postoperative ophthalmologic evaluation following CSF leak repair. OSA was another important risk factor associated with intracranial hypertension. Although 32% of our patients had OSA per medical records, the severity of OSA was not elicited or confirmed by polysomnography as part of the study protocol. Future investigation will be directed toward elucidating any correlation between OSA severity and CSFP changes in this patient population. Continuous CSFP monitoring has the potential to provide useful information for managing patients undergoing the repair of spontaneous CSF leaks. Prior studies assessed CSFP at single time points, which did not take into account ICP changes due to patient positioning or physiologic CSFP changes overnight.11, 12 The continuous monitoring approach is unique in that it enables ongoing detection of abnormal pressure elevations and correlates these changes with physiologic events in real-time. Figure 2 displays a sample recording for a patient with OSA. As the respiratory rate decreased around 12:13 AM (onset of an apnea episode), the ICP increased accordingly followed by a decline in oxygen saturation from 100% to approximately 75%. After the patient regained normal breathing pattern,
ICP lowered to baseline with a return of oxygen saturation. Simultaneous capturing of CSFP and other physiologic measurements offers a dynamic representation of changes in ICP over the patient’s postoperative course. This provides our neurologists with useful clinical information to determine whether additional treatment is required to manage a patient’s ICP after surgical repair. The CSFP measured in this study were lower than those reported in other studies. Unlike that in Illing et al.17 (mean ICP: 27.7 cmH2 O) and Schlosser et al.10 (mean ICP: 32.5 cmH2 O), the pressures in our study were mostly less than 25 cmH2 O. As mentioned previously, opening pressures can be artificially elevated due to concurrent Valsalva maneuvers or other physiologic changes at the time of lumbar drain placement. For this reason, opening pressures were not routinely recorded in our study. Although many patients actually had high postoperative CSFP in the initial recording period, their pressures gradually leveled off throughout the night. Interestingly, another study that used continuous CSFP monitoring similar to our technique also reported a mean postoperative CSFP less than 25 cmH2 O (mean ICP: 15 cmH2 O).18 In addition, continuous CSFP monitoring provides a way to compare preoperative and postoperative CSFP parameters: mean CSFP, peak CSFP, PWA, and AUC 25. In our study, most CSFP parameters increased, whereas mean CSFP and PWA decreased slightly (mean CSFP change: −1.17 cmH2 O or −0.86 mmHg, p = 0.531; peak CSFP change: 1.56 cmH2 O or 1.15 mmHg, p = 0.798; PWA change: −0.83 cmH2 O or −0.61 mmHg, p = 0.183). However, the preoperative and postoperative changes were statistically insignificant. We could not conclude a physiologic increase or decrease of CSFP parameters based on these differences. Rather, they might reflect that CSFP did not alter significantly with the repair or at least directly after the repair. Alternatively, there may not have been enough time for sufficient CSFP elevation to develop after the immediate postoperative period when the pressure measurements were taken. These findings were consistent with studies that showed persistent ICP elevation despite successful repair.9, 10 Although mean CSFP, peak CSFP, and PWA provided useful information on the patient’s ICP
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
76
Continuous CSF pressure monitoring
status, they were single-dimensional pressure variables that did not factor in changes over time. AUC 25 was a timepressure product of CSFP above a certain threshold and is theoretically a more robust way to represent ICP changes. Although the change in AUC 25 was statistically insignificant (p = 0.079), its value increased considerably from 21.8 cmH2 O-min (16.0 mmHg-min) before the repair to 142.2 cmH2 O-min (104.6 mmHg-min) after the repair. Future studies will further assess the validity of AUC measurements in evaluating ICP status. In our practice, continuous CSFP monitoring is used to determine the need for postoperative ICP-lowering therapy to prevent CSF leak recurrences. Acetazolamide decreases CSF production at the choroid plexus and has been effective in reducing ICP via this mechanism.19 Acetazolamide therapy was started only after considering both the patient’s clinical status (presence of signs and symptoms consistent with elevated ICP) and the quantitative CSFP data. In the present study, 8 of 25 (32%) received acetazolamide after the surgery, and 2 patients (8%) underwent ventricular shunt placement (1 had both acetazolamide and a subsequent shunt). Although about one-third of patients required further ICP-lowering therapy, the majority of patients in the study did not need additional interventions. Therefore, prophylactic acetazolamide therapy was not initiated, considering the risks and benefits of starting every patient on this medication. CSFP parameters including mean CSFP, peak CSFP, PWA, and percentage of data above the Dandy criteria were overall higher in patients who received acetazolamide therapy and shunts compared to those who had no further treatment. This was expected given that patients receiving additional ICP-lowering therapies typically showed persistent ICP elevation after their surgery. Of note, higher doses of acetazolamide (1 to 2 g/day) were used for treatment compared to the commonly described dose. If the patient failed to respond to medical therapy, ventricular shunt was considered to further reduce ICP. All study patients were followed postoperatively in our outpa-
tient clinics with an average follow-up time of 526 days. Despite the high recurrence rate reported in the literature, no patients in this study developed a recurrent leak during the study period.3, 4, 10 The strengths and limitations of this study should be considered. The current study expanded from our previous cohort of 12 patients to include 25 who were treated in the same tertiary academic center.14 Although this was one of the larger series in the literature, the small number of participants allowed for only limited power in statistical analyses. All of the operations were performed at a single center, which had the advantage of maintaining consistency among surgical and medical approaches but also limited generalizability. In addition, although the monitoring was approached prospectively, the retrospective nature of data collection in this study only allowed for analysis of data that were already available. Due to a lack of comparison groups, it would be difficult to assess whether the high success rate in our patients was due to CSFP-lowering therapeutic interventions, as determined by the monitoring parameters, or to surgical techniques.
Conclusion In summary, continuous CSFP monitoring may provide clinically important information for managing patients who undergo endoscopic CSF leak repair surgery. It enables providers to correlate physiologic events with otherwise undetected changes in ICP over time and may aid in clinical decision-making for prescribing additional ICP-lowering therapies. Although prospective data with a control group is needed, this study nevertheless demonstrates the potential usefulness of the continuous CSFP monitoring approach. Identification of individuals with persistent ICP elevations after surgery through perioperative pressure monitoring may help in improving patient outcomes and surgical success in patients undergoing spontaneous anterior skull-base CSF leak repair.
References 1. 2.
3.
4.
5.
6.
7.
77
Ommaya AK. Cerebrospinal fluid rhinorrhea. Neurology. 1964;14:106–113. O’Connell JE. Primary spontaneous cerebrospinal fluid rhinorrhea. J Neurol Nerosurg Psychiatry. 1964;27:241–246. Yang Z, Wang B, Wang C, Liu P. Primary spontaneous cerebrospinal fluid rhinorrhea: a symptom of idiopathic intracranial hypertension? J Neurosurg. 2011;115:165–170. Chaaban MR, Illing E, Riley KO, Woodworth BA. Spontaneous cerebrospinal fluid leak repair: a five-year prospective evaluation. Laryngoscope. 2014;124:70–75. Brainard L, Chen DA, Aziz KM, Hillman TA. Association of benign intracranial hypertension and spontaneous encephalocele with cerebrospinal fluid leak. Otol Neurotol. 2012;33:1621–1624. Suryadevara AC, Fattal M, Woods CI. Nontraumatic cerebrospinal fluid rhinorrhea as a result of pseudotumor cerebri. Am J Otolaryngol. 2007;28:242–246. Perez MA, Bialer OY, Bruce BB, Newman NJ, Biousse V. Primary spontaneous cerebrospinal fluid leaks and idiopathic intracranial hypertension. J Neuroophthalmol. 2013;33:330–337.
8.
9.
10.
11.
12.
13.
14.
Shaw GY, Million SK. Benign intracranial hypertension: a diagnostic dilemma. Case Rep Otolaryngol. 2012;2012:814696. Schlosser RJ, Wilensky EM, Grady MS, Bolger WE. Elevated intracranial pressures in spontaneous cerebrospinal fluid leaks. Am J Rhinol. 2003;17:191–195. Schlosser RJ, Wilensky EM, Grady MS, Palmer JN, Kennedy DW, Bolger WE. Cerebrospinal fluid pressure monitoring after repair of cerebrospinal fluid leaks. Otolaryngol Head Neck Surg. 2004;130:443– 448. Poca MA, Sahuquillo J, Topczewski T, Lastra R, Font ML, Corral E. Posture-induced changes in intracranial pressure: a comparative study in patients with and without fluid blocks at the craniovertebral junction. Neurosurgery. 2006;58:899–906. Kropyvnytskyy IV, Saunders FW, Klemfuss H. Circadian rhythm of cerebral perfusion pressure and intracranial pressure. Brain Inj. 1999;13:45–52. Ng I, Lim J, Wong HB. Effects of head posture on cerebral hemodynamics: its influences on intracranial pressure, cerebral perfusion pressure, and cerebral oxygen. Neurosurgery. 2004;54:593–597, discussion 598. Reh DD, Gallia GL, Ramanathan M, et al. Perioperative continuous cerebrospinal fluid pressure monitor-
International Forum of Allergy & Rhinology, Vol. 5, No. 1, January 2015
15.
16.
17.
18.
19.
ing in patients with spontaneous cerebrospinal fluid leaks: Presentation of a novel technique. Am J Rhinol Allergy. 2010;24:238–243. Hegazy HM, Carrau RL, Synderman, CH, Kassam A, Zweig J. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope. 2000;110:1166–1172. Steffen H, Eifert B, Aschoff A, Kolling GH, Volcker ¨ HE. The diagnostic value of optic disc evaluation in acute elevated intracranial pressure. Ophthalmology. 1996;103:1229–1232. Illing E, Schlosser RJ, Palmer JN, Cur´e J, Fox N, Woodworth BA. Spontaneous sphenoid lateral recess cerebrospinal fluid leaks arise from intracranial hypertension, not Sternberg’s canal. Int Forum Allergy Rhinol. 2014;4:246–250. Ramakrishnan VR, Suh JD, Chiu AG, Palmer JN. Reliability of preoperative assessment of cerebrospinal fluid pressure in the management of spontaneous cerebrospinal fluid leaks and encephaloceles. Int Forum Allergy Rhinol. 2011;1:201–205. Lee AG, Wall M. Papilledema: are we any nearer to a consensus on pathogenesis and treatment? Curr Neurol Neurosci Rep. 2012;12:334–339.
