Search tips
Search criteria 


Logo of jnnpsycJournal of Neurology, Neurosurgery and PsychiatryVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Neurol Neurosurg Psychiatry. 2007 October; 78(10): 1149–1151.
PMCID: PMC2117574

Pneumoencephaly following lumbar puncture in association with an ethmoidal osteoma and porencephalic cyst


A 50‐year‐old woman developed pneumoencephaly following a CSF examination for evaluation of dysequilibrium. Previous investigations had demonstrated a number of high signal T2 lesions on MRI of the brain. In addition, there was what was thought to be an asymptomatic cystic lesion in the left frontal lobe communicating with the lateral ventricle. After the lumbar puncture she developed extensive pneumoencephaly with pressure dilatation of the ventricular system. There was CSF rhinorrhoea. Further CT scans showed an osteoma in the ethmoidal air sinus with protrusion into the cystic area. This was the site of both the CSF leak and air entry. Caution must be taken when considering a CSF examination in the presence of either a presumed asymptomatic porencephalic cyst or ethmoid osteoma.

Case report

A 56‐year‐old female presented with a 2 year history of disequilibrium. This was intermittent with episodes lasting up to 2 weeks' duration. She described subjective swaying “as if I just came off a boat”. There was no associated vertigo. She remained asymptomatic between these attacks. Apart from saccadic intrusion on pursuit eye movements, her examination, including Dix Hallpyke's manoeuvre, was normal. She underwent MRI of the brain.

MRI demonstrated multiple subcortical lesions which gave a high signal on both T2 weighted and fluid attenuated inversion recovery (FLAIR) sequences (fig 11).). In addition, there was a cystic lesion in the left frontal lobe with signal characteristics of CSF (fig 22).). It was felt this communicated with the non‐dilated horn of the left lateral ventricle.

figure jn97998.f1
Figure 1 Initial fluid attenuated inversion recovery (FLAIR) imaging demonstrating subcortical high signal change.
figure jn97998.f2
Figure 2 MRI demonstrating left frontal lobe cyst.

In order to evaluate the high signal changes, she was further investigated with a series of blood tests and CSF examination. The CSF examination was performed as a day case in the left lateral position. This was uncomplicated with an opening pressure of 8.5 cm H2O. Having been discharged on the same day after a 4 h period of observation in the ward, she developed sudden onset of severe generalised headache associated with nausea. There was no postural variation in the headache complex. She was admitted and rested on the assumption that this was probably low pressure headache. Over the next 24 h she began to deteriorate with confusion, intermittent dysphasia and mild left‐sided weakness. An urgent CT scan of the brain was performed.

This demonstrated the presence of extensive pneumocephaly with pressure dilation of the ventricular system (fig 33).). She was treated with bed rest and advised not to blow her nose. A nasal discharge was confirmed to be CSF.

figure jn97998.f3
Figure 3 CT demonstrating extensive pneumocephaly.

She was observed, given adequate fluid and bed rest. Over 2 weeks her condition stabilised and she gradually improved. A repeat scan showed near complete resolution of the air in the subarachnoid space, although there was still air in both lateral ventricles and the left frontal cyst. It was therefore felt that she had a CSF fistula and further investigations were undertaken.

T2 weighted MRI showed high signal change within the left anterior ethmoidal sinus air cell compatible with a CSF leak (fig 44).). Fine cut CT imaging with bony window settings demonstrated an osteoma in the ethmoidal air cells, previously unsuspected on MRI (fig 55).). This appeared to protrude up into the cystic area.

figure jn97998.f4
Figure 4 CSF in ethmoidal air cells.
figure jn97998.f5
Figure 5 Osteoma in ethmoidal air cells.

Four weeks after the onset of pneumoencephaly, she underwent repair of her CSF leak and removal of the osteoma. A left frontal craniotomy was performed. The floor of the left frontal fossa was explored where a intradural osteoma arising from the ethmoid area was seen. There was no dura above it and there was an associated dural fistula and brain herniation in the region. Once this was dissected the osteoma was drilled and removed.

The hole was then identified and repaired using a combination of Tisseal, glue and dural patch. Haemostasis was then secured and the dura was repaired in the usual fashion.

Further imaging confirmed successful repair and gradual improvement with absorption of intracranial air and resolution of her symptoms. Her CSF constituents were normal with no evidence of an inflammatory process. It was felt that the high signal changes on T2 weighted imaging were probably vascular in origin and that her dysequilibrium was vestibular.


It is well established that spontaneous pneumoencephaly can occur as a result of osteomas.1,2,3,4,5,6,7,8 Our patient underwent a straightforward, atraumatic, first pass lumber puncture, with constant CSF flow throughout. At no point, either during or after the procedure, was it apparent that air had entered via the lumbar puncture site.

The site of the CSF leak correlated anatomically with that of the ethmoidal osteoma.

The precise method of formation of CSF fistula and pneumocephaly is unclear. It may be in the presence of CSF leakage, the escaping fluid through the defect, be it dural or arachnoidal, draws air by a mechanism comparable with that of a siphon into the relatively low pressure intracranial space. Alternatively, the breached dura acts as a simple flap valve, and air will enter the cranium whenever ambient pressure exceeds that of intracranial pressure. In our case we postulate that the lumbar puncture resulted in the creation of a negative pressure system and that this perpetuated the inflow of air into the cranium.

We do not have an adequate explanation for the presumed porencephalic cyst but postulate that a previous leak may have occurred at some time. The porencephalic cyst was of CSF intensity and had smooth walls. No abnormal signal was seen in the adjacent brain parenchyma. No mass effect or volume loss was appreciated. In view of these findings, the porencephalic cyst is likely to have occurred in early life, probably in utero (congenital). Post traumatic sequelae would be associated with adjacent gliosis and haemorrhagic residua. Such cysts would have irregular margins, unlike what was seen in this case.

In conclusion, a CSF examination may be contraindicated in the presence of an ethmoidal osteoma, as pressure effects may result in an iatrogenic pneumocephaly. Also, incidental frontal porencephalic cysts should be further investigated for the presence of an osteoma which may be missed on standard MRI and even CT scans.9 A high index of suspicion is necessary as consequences could be grave if not treated.


Competing interests: None.


1. Gezici A R, Okay O, Ergun R. et al Rare intracranial manifestations of frontal osteomas. Acta Neurochir (Wien) 2004. 146393–396.396 [PubMed]
2. Johnson D, Tan L. Intraparenchymal tension pneumatocele complicating frontal sinus osteoma: case report. Neurosurgery 2002. 50879–879.879
3. Bramley D C, Ghosh S. Tension pneumocephalus attributable to an ethmoid osteoma presenting as a stroke in evolution: an unusual presentation. Emerg Med J 2001. 18317–318.318 [PMC free article] [PubMed]
4. Huneidi A H, Afshar F. Chronic spontaneous tension pneumocephalus due to benign frontal sinus osteoma. Br J Neurosurg 1989. 3389–392.392 [PubMed]
5. Hardwidge C, Varma T R. Intracranial aeroceles as a complication of frontal sinus osteoma. Surg Neurol 1985. 24401–404.404 [PubMed]
6. Fucci G, Elefante R, Smaltino F. Pneumocephalus secondary to fronto‐ethmoidal osteoma. Radiol Med (Torino) 1984. 70187–189.189 [PubMed]
7. Bartlett J R. Intracranial neurological complications of frontal and ethmoidal osteomas. Br J Surg 1971. 58607–613.613 [PubMed]
8. Reid W L. Spontaneous intracerebral aerocele associated with osteoma of posterior wall of frontal sinus. Med J Aust 1966. 1352–353.353 [PubMed]
9. Brunori A, de Santis S, Bruni P. et al Life threatening intracranial complications of frontal sinus osteomas: report of two cases. Acta Neurochirurg 1996. 1381426–1430.1430

Articles from Journal of Neurology, Neurosurgery, and Psychiatry are provided here courtesy of BMJ Publishing Group