Secondary Logo

Journal Logo

Tonsillar Herniation After Lumbar Puncture in Idiopathic Intracranial Hypertension

Borire, Adeniyi A. MBBS; Hughes, Andrew R. MBBS, FRACP; Lueck, Christian J. PhD, FRACP, FRCP(UK)

doi: 10.1097/WNO.0000000000000239
Clinical Observation

Abstract: A 30-year-old woman with coexisting renal tubular acidosis and idiopathic intracranial hypertension (IIH), treated with acetazolamide, experienced coning (cerebellar tonsillar herniation) after a lumbar puncture (LP). Brain magnetic resonance imaging at initial diagnosis of IIH showed minor tonsillar descent and computed tomographic venography revealed hypoplasia of the left transverse sinus. The patient previously had three uneventful LPs, all of which showed high opening pressures and normal cerebrospinal fluid composition. In retrospect, it was noted that her serum bicarbonate had fallen to 9 mmol/L (normal: 22–28 mm/L) 1 week before the LP. We hypothesize that the combination of cerebral edema (due to worsening metabolic acidosis), poor venous drainage, and preexisting minor tonsillar descent contributed to her post-LP coning.

Department of Neurology (AAB, ARH, CJL), The Canberra Hospital, Canberra, Australia; and The Australian National University Medical School (AAB, ARH, CJL), Canberra, Australia.

Address correspondence to Adeniyi A. Borire, MBBS, Department of Neurology, The Canberra Hospital, PO Box 11, Woden, Canberra, ACT 2611, Australia; E-mail:

The authors report no conflicts of interest.

Lumbar puncture (LP) is indicated in all patients with idiopathic intracranial hypertension (IIH) for the purposes of diagnosis and is performed on many patients, often repeatedly, as a form of treatment (1). Minor descent (less than 5 mm) of the cerebellar tonsils through the foramen magnum is well described (2–6). Despite this, significant cerebellar tonsillar herniation resulting in compromise of brain function (“coning”) is extremely rare; indeed, we are aware of only 1 previous case report of coning in a patient with IIH (7). We report a patient with IIH who coned after a LP. In retrospect, severe metabolic acidosis may have contributed to this life-threatening complication.

Back to Top | Article Outline


A 30-year-old woman was evaluated in our neurology clinic with a 3-month history of bifrontal and periorbital headache, associated with nausea and neck pain. Her headache was worse in the morning, on lying and on bending over. She also reported transient visual obscurations and nocturnal pulsatile tinnitus. Her medical history consisted of idiopathic distal renal tubular acidosis for which she was on sodium bicarbonate 3.36 g/d and potassium citrate 4.32 g/d.

The patient was morbidly obese with a BMI of 48.9 kg/m2. Neurologic examination was unremarkable apart from bilateral papilledema. Visual acuity was normal but visual field testing showed enlarged blind spots and mild concentric field reduction bilaterally. On optical coherence tomography (CT), there was marked increase in the peripapillary retinal nerve fiber layer (RNFL) thickness (right eye: 280 μm; left eye 356 μm). Computed tomography (CT) of the brain was normal, whereas magnetic resonance imaging (MRI) showed a partially empty sella and minor tonsillar descent (4 mm below the foramen magnum). LP performed under fluoroscopy revealed an opening pressure (OP) of 45 cmH2O with normal CSF analysis (1 WBC, glucose 2.8 mmol/L, protein 182 mg/L). CT venography showed stenosis of the right transverse sinus and hypoplasia of the left transverse sinus (Fig 1).

FIG. 1

FIG. 1

The patient was prescribed acetazolamide 1 g/d with mild upward titration of her sodium bicarbonate and potassium citrate. She continued to have headaches over the next 3 months despite achieving an 11 kg weight loss. There was resolution of her papilledema, both clinically and on OCT (peripapillary RNFL thickness: right eye: 112 μm; left eye, 114 μm). Acetazolamide dose was not increased beyond 1.25 g/d because of the risk of worsening her metabolic acidosis. Two additional LPs were performed as temporizing measures to treat her headaches while awaiting review by the neuroradiology service for consideration of venous sinus stenting. Her OPs were persistently high but other CSF studies remained unremarkable.

Eight months later, the patient presented to the emergency department with a 1-week history of worsening headache. She had no visual symptoms but her papilledema had returned bilaterally. LP was repeated (single pass) using a 24 G Sprotte spinal needle, and OP was 43 cmH2O. Twenty milliliters of CSF were removed resulting in immediate improvement of her headache. Seven hours later, the patient developed severe (10/10) global headache with excruciating neck pain. She was unable to lie flat, her back was arched and her neck was held in extreme extension. She subsequently developed severe hypotension, reduced level of consciousness, anisocoria, and bilateral extensor plantar responses. An urgent CT brain showed diffuse cerebral edema with transtentorial herniation and significant tonsillar descent (Fig 2). There was no change on CT venography.

FIG. 2

FIG. 2

The patient was intubated and transferred to the intensive care unit. An external ventricular drain and intracranial pressure (ICP) monitor were inserted. Her ICP continued to rise to more than 70 cmH2O. Accordingly, she underwent urgent bifrontal craniectomy. CSF analysis on this occasion demonstrated 7 white cells (1 polymorphonuclear leukocyte, 6 mononuclear cells). CSF cytology and viral polymerase chain reaction (herpes simplex, varicella zoster, enterovirus) were negative. Her blood count, renal and liver function tests, sepsis screen (blood culture, urine culture, chest x-ray), and inflammatory markers were all normal. It was noted that her serum bicarbonate had fallen to 9 mmol/L (normal: 22–28 mmol/L) on routine testing 1 week previously. Her metabolic acidosis worsened at the time of her collapse, her arterial pH reaching a nadir of 7.19 (normal: 7.35–7.45) with an arterial pCO2 of 19 mm Hg (normal: 35–45 mm Hg). She gradually recovered with no major neurological sequelae apart from some short-term memory loss and impulsivity. Future plans for her care include a shunting procedure and replacement of the cranial bone flap.

Back to Top | Article Outline


Minor tonsillar descent, defined as displacement of the cerebellar tonsils <5 mm into the upper cervical spinal canal, is well described in IIH. However, data on its prevalence are sparse and contradictory. In one case series of 43 patients with IIH, 28% of patients were reported to have mild tonsillar descent (2–4 mm) on MRI (4). This was similar to the 24% found in another study of 68 IIH patients (5), yet substantially higher than the 2.7% based on pretreatment MRI of 74 patients (6). Significant tonsillar herniation or coning remains very rare after LP (8). This complication occurs when there is significant displacement of the posterior fossa contents into the upper cervical canal leading to compression of the lower brainstem and upper cervical cord, CSF outflow obstruction, severe intracranial hypertension, and cerebral edema.

We could find only 1 case report of coning in a patient with IIH, which proved to be fatal (7). A 27-year-old woman previously diagnosed with IIH, presented with photophobia, neck stiffness, blurred vision, and scintillating scotoma while being treated for pneumonia. She underwent 2 lumbar punctures several hours apart. After the first LP (OP 45 cmH2O), she demonstrated signs of coning including neck hyperextension, globally brisk reflexes, and bilateral extensor plantar responses. A second LP was performed (OP <25 cmH2O) and, several minutes after this, she developed respiratory arrest, which led to her death. Autopsy was consistent with coning. Four years earlier, she had experienced respiratory arrest within minutes after an LP, and minor descent of the cerebellar tonsils was noted on her brain MRI.

The risk of coning in patients with IIH after LP was prospectively studied by Paruchuri et al (8). None of the 55 patients in their case series had significant tonsillar herniation (coning). They concluded that LP was safe in IIH but that caution should be applied in patients with severe neck pain exacerbated by movement, focal neurological signs, and minor tonsillar descent. The reason why coning is rare in IIH remains unknown. One hypothesis is that patients with IIH have a gradual reduction in brain compliance because of a combination of increased cerebral blood flow and mild diffuse interstitial edema (9–11). The subacute nature of IIH allows time for the brain to stiffen, thereby reducing its compliance (12). This may also explain why the ventricles do not enlarge despite chronically elevated CSF pressures within the ventricular system. They appear normal or slit-like on neuroimaging in contrast to other processes with similarly high intraventricular pressure. The difference may relate to the speed of onset of the conditions: a more rapid onset does not permit changes in periventricular brain tissue compliance, resulting in ventricular enlargement and periventricular edema (13,14).

Our patient re-presented with her usual headache pattern. Brain CT scan was not ordered before her LP because she had a normal scan 1 month before her admission and three previous LPs without incident. In retrospect, her metabolic acidosis, due to distal renal tubular dysfunction, had worsened with a sharp drop in her serum bicarbonate despite increased doses of sodium bicarbonate and potassium citrate. One neurological manifestation of acidosis is vasogenic cerebral edema, which results from increased cerebral blood flow due to cerebral vasodilation (15). Respiratory acidosis is a more common cause of cerebral edema than metabolic acidosis because of the ability of CO2 to cross the blood–brain barrier easily, but both CO2 and hydrogen ions can cause cerebral vasodilation. Inadequate treatment of our patient's renal tubular acidosis combined with the use of acetazolamide most likely was responsible for the abrupt worsening of her metabolic acidosis, resulting in cerebral edema.


Category 1: a. Conception and design: A. A. Borire, A. R. Hughes, C. J. Lueck; b. Acquisition of data: A. A. Borire, A. R. Hughes, C. J. Lueck; c. Analysis and interpretation of data: A. A. Borire, A. R. Hughes, C. J. Lueck; Category 2: a. Drafting the manuscript: A. A. Borire, A. R. Hughes, C. J. Lueck; b. Revising it for intellectual content: A. A. Borire, A. R. Hughes, C. J. Lueck; Category 3: a. Final approval of the completed manuscript: A. A. Borire, A. R. Hughes, C. J. Lueck.

Back to Top | Article Outline


1. Friedman DI, Liu GT, Digre KB. Revised diagnostic criteria for the pseudotumor cerebri syndrome in adults and children. Neurology. 2013;81:1159–1165.
2. Graber JJ, Racela R, Henry K. Cerebellar tonsillar herniation after weight loss in a patient with idiopathic intracranial hypertension. Headache. 2010;50:146–148.
3. Brew BJ, Garrick R, Connelley TJ. Lumboperitoneal shunting as a cause of visual loss in benign intracranial hypertension. Clin Exp Neurol. 1987;23:233–235.
4. Aiken AH, Hoots JA, Saindane AM, Hudgins PA. Incidence of cerebellar tonsillar ectopia in idiopathic intracranial hypertension: a mimic of the Chiari I malformation. AJNR Am J Neuroradiol. 2012;33:1901–1906.
5. Banik R, Lin D, Miller NR. Prevalence of Chiari I malformation and cerebellar ectopia in patients with pseudotumor cerebri. J Neurol Sci. 2006;247:71–75.
6. Sinclair N, Assaad N, Johnston I. Pseudotumour cerebri occurring in association with the Chiari malformation. J Clin Neurosci. 2002;9:99–101.
7. Sullivan HC. Fatal tonsillar herniation in pseudotumor cerebri. Neurology. 1991;41:1142–1144.
8. Paruchuri SRA, Lawlor M, Kleinhomer K, Mason L, Johnson C. Risk of cerebellar tonsillar herniation after diagnostic lumbar puncture in pseudotumor cerebri. Anesth Analg. 1993;77:398–410.
9. Milhorat T. Classification of the cerebral edemas with reference to hydrocephalus and pseudotumor cerebri. Childs Nerv Syst. 1992;8:301–306.
10. Jefferson A, Clark J. Treatment of benign intracranial hypertension by dehydrating agents with particular reference to the measurement of blind spot area as a means of recording improvement. J Neurol Neurosurg Psychiatry. 1976;39:627–639.
11. Connolly MB, Farrell K, Hill A, Flodmark O. Magnetic resonance imaging in pseudotumor cerebri. Dev Med Child Neurol. 1992;34:1091–1094.
12. Salman M. Benign intracranial hypertension or communicating hydrocephalus: factors in pathogenesis. Med Hypotheses. 1997;49:371–373.
13. Salman M. Why does tonsillar herniation not occur in idiopathic intracranial hypertension? Med Hypotheses. 1999:53;270–271.
14. Bejjani GK. Association of the adult Chiari malformation and idiopathic intracranial hypertension: more than a coincidence. Med Hypotheses. 2003:60;859–863.
15. Biller J, Ferro JM. Neurologic aspects of systemic disease, part I. Handb Clin Neurol. 2014;119:375–377.
© 2015 by North American Neuro-Ophthalmology Society