Article Text

Pneumocephalus as a complication of metastases and eroding infection in the sacral region
  1. Department of Neurology, School of Medicine University of Rijeka, Clinical Hospital Centre, Rijeka, Kresimirova 42, 51000 Rijeka, Croatia
  1. Professor Ksenija Willheim, Department of Neurology, Clinical Hospital Centre, Rijeka, Kresimirova 42, 51000 Rijeka, Croatia.

Statistics from

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.

Pneumocephalus is a well known and rare disease that denotes a pathological presence of intracranial air. Below, we describe an extremely rare case of the development of pneumocephalus.

A 66 year old woman was admitted to our clinic with headache persisting for the previous two months. The headache had become particularly severe and had been accompanied by vomiting during the previous week. One year before being admitted to our clinic, the patient had undergone brain CT along with other tests and the results were normal. Medical records indicated that 10 years ago the patient had undergone surgical treatment of an adenocarcinoma of the rectum. Since then she had been under an oncologist’s observation. Seven years after surgery, metastases developed in the L5 vertebral body and the sacrum. This was confirmed by radiography of the lumbosacral spine and the sacrum, by radioisotope bone scans, by sacrum CT, and by pathohistological analysis of the biopsy of the sacral region. Despite implemented radiation therapy the general condition of the patient deteriorated. She lost weight and finally became paralysed. Skin and subcutaneous tissue defects appeared in the sacral region and did not heal. Repeated cultures isolated Enterococcus, Enterobacter species, andHafnia alvei. The patient refused the proposed necrotomy; instead she received antibiotic treatment. At admission the patient was extremely thin, subfebrile 37°C; cardiocirculatory and respiratory systems were stable. Findings of the anus praeter naturalis in the abdominal region were normal. In the sacral region a deep seated ulcerative lesion 12×9 cm in size was visible, covered by brown strata, which secreted purulent and bloody substance. A neurological examination indicated that the patient was conscious, that her speech was normal, and that there were no signs of meningeal irritation. Flaccid paraplegia was present together with sensory modalities impaired in the right L4 and S1 dermatome and the left L5, S1, S2, and S3 dermatome and she was incontinent for faeces and urine. Routine laboratory blood tests showed the following increases: erythrocyte sedimentation rate 56 mm/h, ALP 105 U/l, ALT 59 U/l, AST 28 U/l and reduced values of: HGB 113, HCT 0.345, MCHC 26.4 g/l.The findings from other laboratory tests were normal. The fundus oculi examination was normal. Findings from radiography of the paranasal sinuses and the skull were normal except for the presence of air in the third and in the lateral ventricles. The brain CT indicated that the size of the ventricular system was normal and the ventricular system was for the most part filled with very hypodense zones which, according to absorptivity, corresponded to the air. Some air was visible around the brainstem. Such a finding indicates a pneumocephalus (fig 1). The sacrum CT showed a large skin and subcutaneous defect. The sacrum was mostly destroyed and filled with zones corresponding to soft mass density. Sporadically, inside the spinal canal some small hypodense zones were visible, the absorptivity of which corresponded to air (fig2). Abdominal and pelvic CT scans were normal. Findings of lung and heart radiography and ECG were normal. The EEG indicated slow activity. Lumbar puncture was not carried out. Because of the condition of the patient neurosurgical intervention did not take place. The patient’s condition gradually deteriorated and she died three days after admission to hospital. The basic pathophysiological mechanism of pneumocephalus is the development of a dural fistula which permits the flowing out of the CSF and the entry of air into the subarachnoidal space, which, because it cannot flow back, successively accumulates intracranially. In head injuries, pneumocephalus is recorded in 0.5% to 1.0% of all cases.1 It may also develop as a complication of neurosurgical treatment of chronic subdural haematoma or after posterior cranial fossa surgery.2 It is described in the craniofacial approach to tumour removal from the base of the anterior cranial fossa as well as in nasal and otological surgery.3 Intracranial air appears as the consequence of eroding otological infections, of eroding tumours of the brain, nose, and nasopharynx in combination with radiotheraphy.4 It is mostly described as a complication which follows insertion of a nasogastric tube and a case has been described also after insertion of a nasopharyngeal tube.5 Subarachnoid-pleural fistula and intracranial air may occur after severe chest trauma. Cases after thoracotomy and surgery of the tumour of the lung apex have also been described.6 Pneumocephalus may also develop when administering epidural anaesthesia and in continuous external lumbar subarachnoid drainage in the treatment of dural fistula.7The first symptom of pneumocephalus is a headache which in anterior cranial fossa fractures or eroding processes may also be accompanied by rhinorrhea. Should the headache intensify and be coupled with other signs of increased intracranial pressure, an urgent brain CT is required to diagnose pneumocephalus. In most cases it is a small quantity of air and the developed fistula may heal spontaneously. Should the fistula persist, urgent operative repair is needed to prevent brain herniation and death.

Figure 1

Brain CT showing intraventricular air.

Figure 2

CT of the sacrum. The arrow points to a small hypodense area representing air.

We consider the case described above to be exceptionally rare given the way pneumocephalus develops. In our opinion, the destruction of the sacrum, of skin, and of subcutaneous tissue of the sacral region, assigned to a malignant process which was hastened by eroding infection, caused a communication between the subarachnoid space and the surface of the skin. We think that the developed sacrodural fistula is the only possible explanation for the development of pneumocephalus.