Secondary Logo

Journal Logo

Original Contribution

Cerebrospinal Fluid Removal for Idiopathic Intracranial Hypertension: Less Cerebrospinal Fluid Is Best

Perloff, Michael D. MD, PhD; Parikh, Simy K. MD; Fiorito-Torres, Franchesca MD; McAdams, Matthew T. MD; Rayhill, Melissa L. MD

Author Information
Journal of Neuro-Ophthalmology: September 2019 - Volume 39 - Issue 3 - p 330-332
doi: 10.1097/WNO.0000000000000759
  • Free


Severe headache is a common symptom of idiopathic intracranial hypertension (IIH), found to be present in more than 90% of patients at the time of diagnosis, with many describing a daily headache (1). A lumbar puncture (LP) can produce immediate headache relief. One study of 80 patients measured headache intensity before and 10–15 minutes after LP, with “normalization” of cerebrospinal fluid (CSF) pressures to 12–17 cm H2O (7). CSF withdrawal to this pressure range produced headache relief in 72% of patients with IIH. However, there can be a drawback to lowering CSF pressure to a “normalized” range. Patients with IIH also can develop debilitating post-LP headaches. The International Classification of Headache Disorders–3 describes this type of headache as occurring within 5 days of the LP, caused by CSF leakage through the dural puncture (2). Female patients are at an increased risk of post-LP headache compared with male patients. In addition, patients with headache the week preceding the LP have higher risk of post-LP headache and longer-lasting, more debilitating post-LP headache, compared with those who did not have preceding headache (3).

The aim of our study was to assess the effect of low-volume CSF removal on IIH-associated and post-LP headaches in a cohort of adults diagnosed with IIH. We hypothesized that patients with IIH undergoing LP may achieve headache relief with a lower incidence of post-LP headache with a closing pressure (CP) goal of 18–20 cm H2O, or “high-normal” pressure.


All patients who underwent LP for IIH, papilledema, or headache as part of routine outpatient care at Boston Medical Center from October 2011 through October 2016 were reviewed. Retrospective data collection and research design were approved by the Boston Medical Center Institutional Review Board. All LPs were performed under fluoroscopic guidance using a 20-gauge Quincke needle with the patient in the prone position, as previously validated (4).

Inclusion criteria were as follows: 1) pre-existing diagnosis of IIH, or opening pressure (OP) and clinical context suspicious for IIH (1); 2) height and weight recorded within 1 year; 3) documented LP data parameters; and 4) patient follow-up to determine if the headache is better, worse, or unchanged. Exclusion criteria were as follows: 1) OP <25 cm H2O without previous diagnosis of IIH and 2) another identified cause of intracranial hypertension.

Post-LP phone calls performed 4–7 days after the procedure (standard Boston Medical Center practice after a fluoroscopy procedure) were performed by a medical assistant who had not seen the patient clinically and asked about the patient's headache compared with pre-LP headache. Follow-up clinic visit records also were reviewed by an evaluator who had not seen the patient to determine whether the patients' headaches were suggestive of relative intracranial hypotension (postural) or whether they were improved, worse, or unchanged. Twelve patients were lost to follow-up, and their data were not included in the analysis of change in headache severity after LP.

Linear regression was used to estimate the association between mean pressure change and volume removed, and between opening pressure and body mass index (BMI) (Microsoft Excel 2013). CP/volume removed was compared between self-reported headache severity (i.e., better, worse, and no change) using analysis of variance (ANOVA). P values of <0.05 (2-sided) were considered statistically significant. ANOVAs were conducted using Social Science Statistics web-based calculators.


Over the 5-year period, 211 patients had an LP for headache, papilledema, or IIH; of those, 146 met inclusion criteria. Most patients were women (n = 144). The mean age was 34.9 years ± 11.0 (standard deviation) years, and the mean BMI was 39.2 ± 10.5 kg/m2 (Fig. 1). Across all IIH patients, the mean OP was 31.7 ± 7.5 cm H2O (n = 146). The mean volume of CSF removed was 9.7 ± 4.6 mL (n = 146). The most common LP level was L2–3 (97 patients), followed by L3–4 (31 patients), L4–5 (14 patients), L1–2 (3 patients), and L5–S1 (1 patient). The CP was not obtained in 9 patients; otherwise, the mean CP was 17.9 ± 2.7 cm H2O (n = 137), matching the target of “high-normal” CP. The mean pressure change (OP−CP) per volume removed was 1.50 ± 0.6 cm H2O/mL (n = 137). These data fit linear regression, but with modest clinical variability to y = 0.92x + 4.82, r2 = 0.38 (Fig. 2A). There was no relationship between OP and BMI (Fig. 2B). During telephone interview the following week, 125 patients had documented headache severity: improved in 64% (80/125), worsened in 26% (33/125), and remained unchanged in 10% (12/125). There was no significant difference in CP/volume removed based on the various outcomes (better, 17.7 ± 2.67 cm H2O/9.2 ± 4.67 mL; same 17.2 ± 4.25/9.1 ± 3.63; worse 18.3 ± 2.24/10.0 ± 4.50; ANOVA: P values 0.38/0.66, respectively). Eleven patients were headache-free, and 11 patients required emergency or hospital care for post-LP headache (Table 1).

FIG. 1.
FIG. 1.:
Distribution of age and BMI of patients with idiopathic intracranial hypertension. BMI, body mass index.
FIG. 2.
FIG. 2.:
A. The relationship between pressure change (OP−CP) and volume of CSF removed, fitted with linear regression. B. No correlation is found comparing opening pressure to BMI. BMI, body mass index; CP, closing pressure; CSF, cerebrospinal fluid; OP, opening pressure.
Assessment of headache in patients with IIH 1 week after lumbar puncture (n = 137)


There is no specific consensus on how much CSF should be removed during diagnostic or therapeutic LPs for IIH. Clinicians often use “high-volume” CSF removal, up to 30 mL in patients suspected of having IIH (5), but there are little data on how CSF volume removal affects change in CSF pressure. One study of patients aged younger than 22 years suggests a 1:1 linear relationship between CSF volume removed and the amount of pressure relieved, when the maximum change in pressure was under 15 cm H2O (6). A similar relationship has not yet been defined in an all-adult population, and there were little data on the CP range recommended to reduce post-LP headaches in patients with IIH. The current study helps address this gap in the literature. Low-volume CSF removal to approximately 18 cm H2O in the setting of IIH seemed to offer improved or stable headache severity. It should also be noted that pressure measurements at singular time points may not reflect average pressures measured over longer periods (2). Typically, about 1/3 of patients experience some low-pressure headache after an LP, with 25% having moderate or severe headaches (general adult population) (3). However, low-pressure headaches are more common in younger patients and women, where 30%–40% of patients may get moderate-severe, low-pressure headaches (3). Our study found that CSF volume removal targeting a “high-normal” CP pressure of 18–20 cm H2O may provide relief of IIH-associated headache in most patients (80/125 patients) and minimize post-LP headache (11/125 patients requiring hospital care).

Our results suggest that for every 1 mL of CSF removed, the CP decreases approximately 1.5 cm H2O (Fig. 2A). Given some variability in our data and to avoid low CP, one strategy that practitioners could use is to first remove 0.5 mL for every 1 cm H2O reduction desired and then recheck pressure to guide further CSF removal. We also demonstrated that there was no correlation between OP and BMI (Fig. 2B), similar to previous studies (7). In addition, L2–3 was the most common level of LP performed under fluoroscopic guidance; this is likely due to less adipose tissue and degenerative disease present at this level compared with the distal lumbar spine.

The generalizability of our findings is limited by being a single-center retrospective observational study and modest sample size. Despite these limitations, this study is one of the largest to date describing the relationship between volume removed and CP in adults diagnosed with IIH. There seems to be a clinically variable 2:3 linear relationship between milliliters of CSF volume removed and CSF pressure change in centimeters H2O within our population's range of CSF pressures. CSF volume removal targeting a “high-normal” CP pressure of 18–20 cm H2O may provide relief of IIH-associated headache and may avoid worsening post-LP headache for most patients, although future multicenter prospective comparative studies are needed.


M. D. Perloff, S. K. Parikh, F. Fiorito-Torres, M. T. McAdams, and M. L. Rayhill: study concept and design; analysis and interpretation of data; and critical revision of manuscript for intellectual content: M. D. Perloff: study supervision and procedure development: S. K. Parikh, F. Fiorito-Torress, M. T. McAdams, and M. L. Rayhill: acquisition of data.


1. Wall M. The headache profile of idiopathic intracranial hypertension. Cephalalgia. 1990;10:331–335.
2. Headache Classification Committee of the International Headache Society (IHS). The International Classification of Headache Disorders. Vol 38. 3rd edition. London, UK: Sage Publications Ltd; 2018:1–211.
3. Kuntz KM, Kokmen E, Stevens JC, Miller P, Offord KP, Ho MM. Post-lumbar puncture headaches: experience in 501 consecutive procedures. Neurology. 1992;42:1884–1887.
4. Abel AS, Brace JR, McKinney AM, Harrison AR, Lee MS. Practice patterns and opening pressure measurements using fluoroscopically guided lumbar puncture. AJNR Am J Neuroradiol. 2012;33:823–825.
5. Thurtell MJ, Wall M. Idiopathic intracranial hypertension (pseudotumor cerebri): recognition, treatment, and ongoing management. Curr Treat Options Neurol. 2013;15:1–12.
6. McLaren SH, Monuteaux MC, Delaney AC, Landschaft A, Kimia AA. How much cerebrospinal fluid should we remove prior to measuring a closing pressure? J Child Neurol. 2017;32:356–359.
7. Friedman DI, Quiros PA, Subramanian PS, Mejico LJ, Gao S, McDermott M, Wall M. Headache in idiopathic intracranial hypertension: findings from the idiopathic intracranial hypertension treatment trial. Headache. 2017;57:1195–1205.
Copyright © 2019 by North American Neuro-Ophthalmology Society