Article Text

Download PDFPDF

Visual outcomes of surgical intervention for pseudotumour cerebri: optic nerve sheath fenestration versus cerebrospinal fluid diversion
  1. Pedro L Fonseca1,
  2. Daniele Rigamonti2,
  3. Neil R Miller1,2,3,
  4. Prem S Subramanian1,2,3
  1. 1Department of Ophthalmology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  2. 2Department of Neurosurgery, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  3. 3Department of Neurology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  1. Correspondence to Dr Prem S Subramanian, Wilmer Eye Institute, 600 N Wolfe St, Woods 457, Baltimore, MD 21287, USA; psubram1{at}


Objective To compare visual outcomes in pseudotumour cerebri (PTC) patients who underwent either optic nerve sheath fenestration (ONSF) or cerebrospinal fluid (CSF) diversion (shunt) for treatment of visual loss from papilloedema.

Methods Comparative case series of thirty-three patients (33 eyes) who underwent either ONSF (14 patients) or CSF shunt (19 patients) for papilloedema and vision loss. Preoperative and final postoperative visual acuity and mean deviation (MD) on automated perimetry were assessed in addition to papilloedema severity and resolution. Mann-Whitney U test was employed to evaluate differences in final visual outcomes.

Results Postoperative visual acuity did not differ between ONSF (76.4 ETDRS equivalent letters) and shunt (76.4 letters), although there was a trend toward worse preoperative acuity in the ONSF cohort. Final MD was significantly better after shunt (−9.23 dB) compared to ONSF (−17.29 dB), U=52.0, p=0.036. Preoperative papilloedema was qualitatively worse in the ONSF group.

Conclusions Visual acuity and MD improved after ONSF and shunt procedures in patients with severe vision loss from PTC and papilloedema. Improvement of MD in both groups was of the same magnitude (6 dB), but because the two groups were not entirely comparable, prospective comparison of the two procedures is required to establish the relative efficacy of each procedure.

  • Optic Nerve
  • Treatment Surgery
  • Field of vision

Statistics from


Pseudotumour cerebri (PTC), often referred to as idiopathic intracranial hypertension (IIH), is characterised by elevated intracranial pressure (ICP) without clinical or imaging evidence of a space-occupying lesion, and with normal cerebrospinal fluid (CSF) content.1 ,2 PTC may occur at any age, and the cause is unknown. It is postulated that increased intra-abdominal and/or intrathoracic pressure might play a major role in pathogenesis,3 explaining in part a clear predilection for obese young women. In fact, when that specific population is considered, the annual incidence appears to be as high as 15–19 cases per 100 000.4

Patients with PTC may experience transient visual obscurations, dyschromatopsia, reduced contrast sensitivity, loss of central visual acuity, or a combination of these. Additionally, visual field defects are nearly always present,5 but because they are often initially mild-to-moderate, about half the patients with PTC are asymptomatic when first examined.6 These field defects are often reversible, but population-based studies nevertheless have shown that permanent visual impairment, often severe, occurs in over 10% of patients regardless of treatment, and in a long-term, hospital-based study, 24% of patients suffered severe visual loss.7 ,8

There is no universally accepted treatment for PTC. Thus, the treatment a patient receives often depends on the type of specialist who first evaluates her or (rarely) him.9 Generally, less severe cases are treated medically. Severe and medically refractory cases of PTC that are threatening vision, are generally treated surgically. Potential surgical options include optic nerve sheath fenestration (ONSF), CSF diversion procedures10 and, more recently, endovascular stenting for haemodynamically significant venous sinus stenoses.11 Despite extensive anecdotal evidence showing that ONSF and CSF diversion procedures are effective in stabilising or, in many cases, improving visual function, there has been no randomised, prospective clinical trial comparing surgical procedures in PTC,12 and all retrospective series have focused on just one type of intervention. Thus, it is not surprising that a recent Cochrane Collaboration review did not find sufficient data to support the use of one form of treatment rather than another.9 We report our experience on the visual outcomes of PTC patients who underwent ONSF or CSF diversion procedures.

Materials and methods

We performed a retrospective chart review of patients with the diagnosis of PTC who underwent ONSF or a CSF diversion procedure at the Johns Hopkins Hospital (JHH) between January 2001 and December 2011. Institutional review board approval was obtained, and the study was compliant with the Health Insurance Portability and Accountability Act guidelines.

Patients eligible for this study were those with a confirmed diagnosis of PTC fulfilling the Modified Dandy Criteria2 in whom papilloedema and vision loss were not considered amenable to medical treatment, or that did not respond to maximum medical therapy, who therefore, underwent surgical treatment at JHH. The list was obtained from the clinical records of the authors and was supplemented by a search of hospital billing records. Clinical records were obtained, and data from the preoperative and last clinical visit were collected. Timing and type of surgical intervention were recorded. Only patients in whom there were no other causes of visual loss, and who had a postoperative follow-up of at least 3 months were included in the study.

Snellen best-corrected visual acuity (BCVA) was recorded and converted to ETDRS equivalent for the purpose of analysis.13 Visual field assessment was performed using a Humphrey Field Analyser with a 24-2 SITA-standard protocol. The eye with the worse mean deviation (MD) on preoperative visual field assessment was selected. Tests with fixation losses under 50% were considered reliable and used for analysis. Stereo fundus photographs were obtained after pupillary dilation, and graded according to the Frisén papilloedema grade14 by two neuro-ophthalmologists (NRM, PSS) and, when available, the photos were collected during data review. Patients were divided into two groups depending on whether they underwent ONSF or a CSF diversion procedure. Descriptive statistics were computed for the mean, SD, and range for quantitative variables. The Mann–Whitney U test was used to compare preoperative and surgical outcomes in both groups at the two-sided 5% level of significance. SPSS software (V.20, SPSS, Chicago, Illinois, USA) was used for data analysis.


Forty-one patients (38 women and 3 men) with PTC underwent surgical treatment; this cohort represents less than 10% of all PTC patients seen at the Wilmer Eye Institute during the study period. Of these patients, 33 (30 women and 3 men) had postoperative follow-up of at least 3 months, and were included in the study. The age at the time of surgery ranged from 9 to 52 years (mean, 30.9). Mean ages were 25.9 in the ONSF group and 34.6 in the CSF diversion procedure (U=203.5, p=0.009).

Fourteen patients (11 women and 3 men) underwent unilateral (12 patients) or bilateral (2 patients) ONSF. Unilateral ONSF was always performed in the eye with the worse MD on visual fields assessment. Preoperative BCVA in this group was 60.4 ETDRS-equivalent letters (SD, 30.6), improving to 76.4 ETDRS-equivalent letters (SD, 14.2) over a mean postoperative follow-up of 16.8 months (SD, 10.3) (figure 1A). Seven of the 14 patients experienced improvement of at least 5 ETDRS-equivalent letters of visual acuity, whereas the rest had stable visual acuity (four patients) or worse than 5 ETDRS-equivalent letters visual acuity loss (three patients). In this group, 11 patients had reliable visual fields. Before ONSF, all patients had visual field defects, with an average MD of −23.50 dB (SD, 7.45). Postoperative MD was −17.29 dB (SD, 9.90) (figure 2A). Preoperative papilloedema grades were available for 12 patients (85.7%) and was grade 3 in 2 patients, grade 4 in 9 patients and grade 5 in 1 patient. Thus, the median grade of papilloedema in this group was 4. Of the remaining two patients, the papilloedema was simply described as severe in one case and moderate in the other. Papilloedema improved in 11 patients (78.6%) after ONSF, with complete resolution in 10 (71.4%), and there were no cases of overt optic atrophy; however, persistent papilloedema refractory to medical treatment required shunt implantation in three patients (21.4%). There were no surgical complications in any of the 14 patients who underwent ONSF.

Figure 1

Visual acuity (ETDRS equivalent) before and after surgery in patients undergoing optic nerve sheath fenestration (ONSF) (A) or cerebrospinal fluid diversion procedures (eg, ventriculoperitoneal or ventriculoatrial shunting) (B). Error bars show SD.

Figure 2

Mean deviation on standard automated perimetry (Humphrey Field Analyser, 24-2 SITA-Standard protocol) before and after either optic nerve sheath fenestration (ONSF) (A) or shunting (B). Error bars indicate SD.

Nineteen patients (all women) underwent CSF diversion. Ventriculoperitoneal shunt implantation was performed in seven patients, and ventriculoatrial shunts were placed in 12 patients. No patient underwent a lumboperitoneal shunt. Preoperative mean BCVA in this group was 76.3 ETDRS-equivalent letters (SD, 9.6) compared with 80.0 ETDRS-equivalent letters (SD, 7.8) after surgery. Postoperative follow-up in this group was 34.0 months (SD, 21.3) (figure 1B). Visual acuity improved in 9 patients, stabilised in 6 and worsened in 4. Visual fields were reliable in all patients. Mean MD before surgery in this group of patients was −15.80 dB (SD, 12.03), and postoperatively was −9.23 dB (SD, 10.08) (figure 2B). Four of the 19 patients in this group had optic atrophy with no optic disc swelling at the time of their surgery. These four patients all had preoperative visual acuities equal to or greater than 80 ETDRS-equivalent letters, and experienced substantial postoperative visual field improvement. Of the remaining 15 patients, preoperative papilloedema classification was available for 13 (86.7%) and was grade 1 in 3 patients, grade 2 in 6 patients, grade 3 in 3 patients and grade 4 in 1 patient. Thus, median papilloedema grade in this group was 2. In the remaining two patients, papilloedema was described as mild. After surgery, 13 of the 15 patients with preoperative papilloedema experienced complete resolution of optic disc swelling, whereas two patients had persistent grade 1 papilloedema at last follow-up. One patient developed a postoperative shunt infection, and eight patients required shunt revisions motivated by persistent papilloedema (two patients) and persistent headache not amenable to medical treatment (six patients).

Comparison between both types of surgical interventions did not show significant differences in preoperative BCVA (U=163.0, p=0.287), postoperative BCVA (U=145.5, p=0.653) or preoperative visual field MD (U=96.0, p=0.138). Postoperative MD was significantly worse after ONSF than after CSF diversion procedures (U=143.0, p=0.037).


Although patients with PTC may experience loss of vision from haemorrhages, exudates, or subretinal fluid in the macular region, visual loss in PTC usually is caused by progressive optic nerve dysfunction associated with severe papilloedema. Thus, surgical treatment for patients with PTC associated with pre-existing or impending vision loss must reduce the papilloedema rapidly to protect visual function. In this study, we found that ONSF and CSF diversion were effective in reducing papilloedema and improving or stabilising visual function in almost all patients. Numerous studies have demonstrated the efficacy of ONSF15–18 and CSF diversion.19–21 Permanent surgical complications were not seen in our ONSF group, and CSF diversion failures were in accord with previously reported rates.20 ,22 The goals of this study were to evaluate the efficacy of each procedure and to determine if one might be superior in protecting visual function. Although a similar analysis was performed via literature review,23 direct comparative data from a single institution have not been reported.

As with any retrospective study, this study has some limitations. First, baseline visual function in the two groups, although not statistically different, may have influenced the surgical procedure chosen and the final visual outcome. We did not perform subgroup analyses on our cohorts, as information about surgical decision making was not always available in the medical records, and the small number of subjects in each group would have limited the statistical validity of any conclusions. It is possible that the likelihood of visual recovery was influenced by the severity of presurgical vision loss.24 There are no standard criteria defining medical failure,25 nor are there guidelines to indicate when a patient's optic nerve is at increased risk of damage from papilloedema and must be treated with surgery. Severe vision loss at presentation may be a poor prognostic indicator and often prompts immediate surgery.7 ,24 In this study, four patients in the shunt group had preoperative optic atrophy, and they all had substantial visual improvement. Since no retinal nerve fibre layer thickness measurements were performed, it is possible these were patients in whom optic atrophy was mild. Although we agree that overt preoperative optic atrophy, in general, is a poor prognostic factor, our results suggest that the simple presence of preoperative optic atrophy does not preclude visual recovery, at least in some cases. In this study, there were five patients (two in the ONSF group and three in the shunt group) who were excluded because of insufficient follow-up. Both patients in the ONSF group had a much worse preoperative visual acuity and worse papilloedema than the three patients in the shunt group and, as such, we do not believe these losses to follow-up would substantially affect the results.

Clinical features, such as how acute the papilloedema appears (presence of haemorrhages, nerve fibre layer infarcts) and patient compliance with medical treatment also influence decision making, and this information was not consistently available in the medical records. Additionally, patients who underwent CSF diversion in this study had lower grades of papilloedema, overall, as well as a trend towards a lesser degree of preoperative visual dysfunction compared with patients in the ONSF group. Thus, it is possible that a clinical bias existed that resulted in patients with more severe papilloedema and/or vision loss undergoing emergent ONSF rather than shunting, as it is often more straightforward to perform the former procedure emergently. Additionally, resolution of papilloedema after ONSF tends to occur within hours or days, whereas the course of improvement after CSF diversion may be slower. No direct comparative studies address this potential difference in rate of resolution. It is rare for papilloedema not to resolve after ONSF, although recurrence can occur, and progressive visual loss despite apparently successful surgery and papilloedema resolution has been described.26 Similarly, vision loss that persists after CSF diversion has been treated with subsequent ONSF with an improved visual outcome after the second surgical procedure.27 As the incidence of PTC has increased in the USA with burgeoning obesity, shunting rates have also climbed.28 ONSF and CSF diversion have a significant long-term failure rate as high as 60%.18–20 CSF diversion is more effective in treating headache,20 but pressure-independent headache also is common in patients with PTC and may require medical treatment despite relief of papilloedema and normalisation of ICP.29

Most patients in this study, as in the PTC population as a whole, had bilateral papilloedema or optic nerve involvement, and we evaluated the worse eye in this analysis. Because this was a retrospective study, we did not believe that the additional information provided by using general estimating equations would add meaningful data to our analysis and would have confounding potential.30 Additionally, clinical decisions about surgery are made to protect vision in the eye in which vision is declining, even if the fellow eye has relatively well-preserved visual function.

In conclusion, our data demonstrate that ONSF and CSF diversion effectively reduce papilloedema and are associated with improved visual function as measured by visual acuity testing and automated perimetry in patients with PTC. Although patients who underwent CSF diversion had a statistically better final outcome on automated perimetry with a trend toward better pretreatment visual function compared with patients who underwent ONSF, these patients had less severe papilloedema overall. It is notable that in both groups, the MD improved by 6 dB (−23.46 to −17.29 dB for ONSF, and −15.77 to −9.23 dB for shunt) at the final postoperative exam. Our retrospective analysis does not permit us to determine if the amount of MD improvement between the groups was significant or not, but it suggests that ONSF does preserve visual field function. The rate of complications and further vision loss were very low in both groups and did not differ statistically. There are theoretical advantages to performing ONSF for severe papilloedema and considering CSF diversion when headache is more severe. The difference in economic impact on patients between ONSF (an outpatient procedure, ie, usually not repeated, but may not relieve headache burden) and CSF diversion (inpatient surgery with a high incidence of repeat surgeries needed over time) could not be determined in this population, but may be another important factor in deciding which procedure to choose. A prospective randomised trial is needed to address definitely these issues.



  • Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Seattle, Washington, May 2013.

  • Contributors PLF contributed to study design, collected the data, performed initial data analyses, and prepared parts of the manuscript. DR provided patient records and data, reviewed the data analyses, and provided critical revision of the manuscript. NRM provided clinical records and data and critical review of the manuscript. PSS provided supervision and funding for the research, contributed to study design, prepared parts of the manuscripts, and provided critical review of the final submission.

  • Funding Supported in part by an unrestricted grant to the Wilmer Eye Institute from Research to Prevent Blindness, USA.

  • Competing interests None.

  • Ethics approval Johns Hopkins University School of Medicine IRB.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost Jonas