Pneumocephalus following ventriculoperitoneal shunt Case report LAWRENCE H. PITTS, M.D., CHARLESB. WILSON, M.D., HERBERT H . DEDO, M . D . , AND ROBERT WEYAND,M . D .
Departments of Neurological Surgery and Otolaryngology, University of California School of Medicine, San Francisco, California v" The authors describe a case of massive pneumocephalus following ventriculoperitoneal shunting for hydrocephalus. After multiple diagnostic and surgical procedures, congenital defects in the tegmen tympani of both temporal bones were identified as the sources for entry of air. A functioning shunt intermittently established negative intracranial pressure and allowed ingress of air through these abnormalities; when the shunt was occluded, air did not enter the skull, and there was no cerebrospinal fluid leakage. Repair of these middle ear defects prevented further recurrence of pneumocephalus. KEY WOROS 9 pneumocephalus 9 hydrocephalus ventriculoperitoneal shunt 9 CSF
9 aural fistula
9
later, evaluation for progressive ataxia revealed communicating hydrocephalus and a VP shunt was placed. At revision 1 year later, intraventricular air was noted (Fig. 1). At this time the patient became demented and intermittently quadriparetic. Examination. Careful evaluation for gasproducing organisms, isotope cisternography, and tomography failed to elucidate the source of entry of gas. At no time was CSF otorrhea or rhinorrhea noted. We could remove the air by way of the ventricular catheter and prevent the recurrence of pneumocephalus by clipping the peritoneal catheter, but each time we C a s e Report opened the shunt pneumocephalus recurred. At the age of 8 years, this patient un- Transsphenoidal packing of the sphenoid derwent resection of a cerebellar astrocytoma sinus and bifrontal craniotomy with packing followed by radiation therapy. Twelve years of the planum sphenoidale and cribriform
N 1884, ChiarP reported the first autopsied case of intracranial pneumocephalus resulting from ethmoiditis. Radiographic evidence of pneumocephalus in a patient with a skull fracture was first presented by Luckett7 in 1913. Since then, many cases of pneumocephalus have been recognized, and at least 350 have been reported. Pneumocephalus following a ventriculoperitoneal (VP) shunt for communicating hydrocephalus has not previously been described. We are reporting such a case.
I
J. Neurosurg. / Volume 43 / November, 1975
631
L. H. Pitts, C. B. W i l s o n , H . H . D e d o a n d R. W e y a n d
FI6. 1. Anteroposterior (left) and lateral (right) skull films showing marked pneumocephalus. Note ventricular catheter tip on the floor of the right frontal horn.
plates did not prevent pneumocephalus when the shunt was opened. Petrous tomograms revealed defects in both tegmen tympani, more pronounced on the left. Operation. The floor of the nose, inferior septum, and inferior turbinate was anesthetized with 5% cocaine on a cotton applicator. Anteroposterior and lateral x-ray films of the skull were taken to establish a baseline for the amount of residual air in the skull. A metal Eustachian catheter was then placed through the nose and into the opening of the Eustachian tube. An air hose with a pistol grip tip was placed part way into the proximal end of the Eustachian catheter, and a listening tube was placed between the patient's ear and the surgeon's ear. Short bursts of air were blown into the Eustachian catheter. Passage of air into the middle ear was confirmed by means of the listening tube. We then verified by x-ray films that air had passed into the cranial vault. Insutflation of air into the left Eustachian tube caused sudden headache and nystagmus; air appeared subdurally along the clivus but the intraventricular air did not increase. Insufflation of the right Eustachian tube caused neither symptoms nor intracranial air. The left subtemporal region was explored and the large congenital bone and dural defects were repaired with temporalis fascia and tissue adhesive, but pneumocephalus still recurred. 632
A simple mastoidectomy was performed to repair the defect in the right tegmen tympani. Postoperative Course. No complications followed this procedure. The patient was already deaf on the right side so that she suffered no further hearing loss. Pneumocephalus has not recurred with a functioning shunt, but the patient remains retarded. Discussion
We present this case report because of the unusual occurrence of pneumocephalus secondary to CSF diversion. Each time we opened the VP shunt, a pressure gradient developed which favored drainage of ventricular fluid and replacement by air through the congenital temporal bone defects. There was no leakage of CSF, presumably because of a ball-valve action of brain tissue or a dural leaflet that covered the defect and allowed ingress of air without escape of fluid. The fistula must have been tenuous since the shunt remained patent and there was no pneumocephalus for several months before the bone defects were repaired. The causes of pneumocephalus are well known. In Markham's series 8 trauma was by far the leading cause; other cases were caused by osteomas of the paranasal sinuses, pituitary tumors both before and after operation, and intracranial abscesses involving gasproducing organisms. Bell, et al.fl reported a
J. Neurosurg. / Volume 43 / November, 1975
Pneumocephalus following VP shunt case of "low spinal fluid pressure syndrome," or " a l i q u o r r h e a , " in which intrathecal pressure was below atmospheric pressure and, consequently, air was sucked inward when the needle stylet was withdrawn during a routine lumbar puncture. Schaltenbrand 1~ reported two similar cases. Kessler and Stern 6 reported a "bronchoventricular fistula" following the revision of a ventriculopleural shunt; they postulated that adhesions of the lung allowed the valveless shunt catheter to abrade the lung surface and erode into a bronchus with retrograde flow of air from lung to ventricle. Intraoperative air insuffiation has been used to identify a subfrontal source of C S F rhinorrhea 9 but we are unaware of the previous use of Eustachian insufflation to locate petrous defects. N o complications followed this procedure in our patient. Finsnes 5 reported a death from intracranial complications following external canal inflation with a pneumatic otoscope for treatment of serous otitis media, and cited other fatal cases that involved either external canal or Eustachian t u b e air insufflation under pressure. The presumed etiology was air leakage through petrous defects to the extradural space and either a pressurized epidural air mass causing severe brain compression or possibly air around the extradural carotid artery causing its occlusion. The risk in Eustachian catheterization and insuttlation can be diminished by: 1) x-ray monitoring for the earliest appearance of intracranial air; 2) using limited quantities of air; and 3) limiting pressures to less than 150 mm Hg. ~ If the usual techniques for localization of C S F or air fistulas are unsuccessful, Eustachian insufflation may prove to be a useful adjunct for locating dural defects. We agree with Dandy 4 that, "disclosure of the fistulous t r a c t m a y be exceedingly difficult, perhaps even impossible." In Ray's 9 series, 27% of 41 patients required more than one operation for repair of C S F leak, and one patient r e q u i r e d five c r a n i o t o m i e s for successful repair. In his series, 10% of the
J. Neurosurg. / Volume 43 / November, 1975
fistulas were not closed. Our patient underwent four operative procedures before closure of the air fistula was accomplished. References I. Ahr6n C, Thulin CA: Lethal intracranial complications following inflation in the external auditory canal in treatment of serious otitis media and due to defects in the petrous bone. Aeta Otolaryngol (Stoekh) 60:407-421, 1965 2. Bell WE, Joynt R J, Sahs AL" Low spinal fluid pressure syndromes. Neurology 10:512-514, 1960 3. Chiari H: Ober einen Fall yon Luftansammlung in den Ventrikeln des menschlichen Gehirns. Z Heilkd 5:383-390, 1884 4. Dandy WE: Treatment of rhinorrhea and otorrhea. Arch Surg 49: 75-85, 1944 5. Finsnes KA: Lethal intracranial complication following air insufflation with a pneumatic otoscope. Acta Otolaryngol (Stock) 75: 436-483, 1973 6. Kessler LA, Stern WZ: Ventriculo-pleural shunt for hydrocephalus. Case report of an unusual complication. J Pediatr 60:418-420, 1962 7. Luckett WH: Air in the ventricles of the brain, following a fracture of the skull. Surg Gyneeol Obstet 17:237-240, 1913 8. Markham JW: The clinical features of pneumocephalus based upon a survey of 284 cases with report of 11 additional cases. Acta Neurochirurgica 16:1-78, 1967 9. Ray BS, Bergland RM: CSF fistula: clinical aspects, techniques of localization and methods of closure. J Neurosurg 30:399-405, 1969 10. Schaltenbrand G: Die akute aliquorrhoe. Verh Dtsch Ges Inn Med 52:473-481, 1940
This work was supported in part by USPHS Training Grant NINDS 5593. This paper was presented in part at the Annual Meeting of the Neurosurgical Society of America, 1973, Bermuda. Address reprint requests to: Charles B. Wilson, M.D., Department of Neurological Surgery, University of California School of Medicine, San Francisco, California 94143.
633
Departments of Neurological Surgery and Otolaryngology, University of California School of Medicine, San Francisco, California v" The authors describe a case of massive pneumocephalus following ventriculoperitoneal shunting for hydrocephalus. After multiple diagnostic and surgical procedures, congenital defects in the tegmen tympani of both temporal bones were identified as the sources for entry of air. A functioning shunt intermittently established negative intracranial pressure and allowed ingress of air through these abnormalities; when the shunt was occluded, air did not enter the skull, and there was no cerebrospinal fluid leakage. Repair of these middle ear defects prevented further recurrence of pneumocephalus. KEY WOROS 9 pneumocephalus 9 hydrocephalus ventriculoperitoneal shunt 9 CSF
9 aural fistula
9
later, evaluation for progressive ataxia revealed communicating hydrocephalus and a VP shunt was placed. At revision 1 year later, intraventricular air was noted (Fig. 1). At this time the patient became demented and intermittently quadriparetic. Examination. Careful evaluation for gasproducing organisms, isotope cisternography, and tomography failed to elucidate the source of entry of gas. At no time was CSF otorrhea or rhinorrhea noted. We could remove the air by way of the ventricular catheter and prevent the recurrence of pneumocephalus by clipping the peritoneal catheter, but each time we C a s e Report opened the shunt pneumocephalus recurred. At the age of 8 years, this patient un- Transsphenoidal packing of the sphenoid derwent resection of a cerebellar astrocytoma sinus and bifrontal craniotomy with packing followed by radiation therapy. Twelve years of the planum sphenoidale and cribriform
N 1884, ChiarP reported the first autopsied case of intracranial pneumocephalus resulting from ethmoiditis. Radiographic evidence of pneumocephalus in a patient with a skull fracture was first presented by Luckett7 in 1913. Since then, many cases of pneumocephalus have been recognized, and at least 350 have been reported. Pneumocephalus following a ventriculoperitoneal (VP) shunt for communicating hydrocephalus has not previously been described. We are reporting such a case.
I
J. Neurosurg. / Volume 43 / November, 1975
631
L. H. Pitts, C. B. W i l s o n , H . H . D e d o a n d R. W e y a n d
FI6. 1. Anteroposterior (left) and lateral (right) skull films showing marked pneumocephalus. Note ventricular catheter tip on the floor of the right frontal horn.
plates did not prevent pneumocephalus when the shunt was opened. Petrous tomograms revealed defects in both tegmen tympani, more pronounced on the left. Operation. The floor of the nose, inferior septum, and inferior turbinate was anesthetized with 5% cocaine on a cotton applicator. Anteroposterior and lateral x-ray films of the skull were taken to establish a baseline for the amount of residual air in the skull. A metal Eustachian catheter was then placed through the nose and into the opening of the Eustachian tube. An air hose with a pistol grip tip was placed part way into the proximal end of the Eustachian catheter, and a listening tube was placed between the patient's ear and the surgeon's ear. Short bursts of air were blown into the Eustachian catheter. Passage of air into the middle ear was confirmed by means of the listening tube. We then verified by x-ray films that air had passed into the cranial vault. Insutflation of air into the left Eustachian tube caused sudden headache and nystagmus; air appeared subdurally along the clivus but the intraventricular air did not increase. Insufflation of the right Eustachian tube caused neither symptoms nor intracranial air. The left subtemporal region was explored and the large congenital bone and dural defects were repaired with temporalis fascia and tissue adhesive, but pneumocephalus still recurred. 632
A simple mastoidectomy was performed to repair the defect in the right tegmen tympani. Postoperative Course. No complications followed this procedure. The patient was already deaf on the right side so that she suffered no further hearing loss. Pneumocephalus has not recurred with a functioning shunt, but the patient remains retarded. Discussion
We present this case report because of the unusual occurrence of pneumocephalus secondary to CSF diversion. Each time we opened the VP shunt, a pressure gradient developed which favored drainage of ventricular fluid and replacement by air through the congenital temporal bone defects. There was no leakage of CSF, presumably because of a ball-valve action of brain tissue or a dural leaflet that covered the defect and allowed ingress of air without escape of fluid. The fistula must have been tenuous since the shunt remained patent and there was no pneumocephalus for several months before the bone defects were repaired. The causes of pneumocephalus are well known. In Markham's series 8 trauma was by far the leading cause; other cases were caused by osteomas of the paranasal sinuses, pituitary tumors both before and after operation, and intracranial abscesses involving gasproducing organisms. Bell, et al.fl reported a
J. Neurosurg. / Volume 43 / November, 1975
Pneumocephalus following VP shunt case of "low spinal fluid pressure syndrome," or " a l i q u o r r h e a , " in which intrathecal pressure was below atmospheric pressure and, consequently, air was sucked inward when the needle stylet was withdrawn during a routine lumbar puncture. Schaltenbrand 1~ reported two similar cases. Kessler and Stern 6 reported a "bronchoventricular fistula" following the revision of a ventriculopleural shunt; they postulated that adhesions of the lung allowed the valveless shunt catheter to abrade the lung surface and erode into a bronchus with retrograde flow of air from lung to ventricle. Intraoperative air insuffiation has been used to identify a subfrontal source of C S F rhinorrhea 9 but we are unaware of the previous use of Eustachian insufflation to locate petrous defects. N o complications followed this procedure in our patient. Finsnes 5 reported a death from intracranial complications following external canal inflation with a pneumatic otoscope for treatment of serous otitis media, and cited other fatal cases that involved either external canal or Eustachian t u b e air insufflation under pressure. The presumed etiology was air leakage through petrous defects to the extradural space and either a pressurized epidural air mass causing severe brain compression or possibly air around the extradural carotid artery causing its occlusion. The risk in Eustachian catheterization and insuttlation can be diminished by: 1) x-ray monitoring for the earliest appearance of intracranial air; 2) using limited quantities of air; and 3) limiting pressures to less than 150 mm Hg. ~ If the usual techniques for localization of C S F or air fistulas are unsuccessful, Eustachian insufflation may prove to be a useful adjunct for locating dural defects. We agree with Dandy 4 that, "disclosure of the fistulous t r a c t m a y be exceedingly difficult, perhaps even impossible." In Ray's 9 series, 27% of 41 patients required more than one operation for repair of C S F leak, and one patient r e q u i r e d five c r a n i o t o m i e s for successful repair. In his series, 10% of the
J. Neurosurg. / Volume 43 / November, 1975
fistulas were not closed. Our patient underwent four operative procedures before closure of the air fistula was accomplished. References I. Ahr6n C, Thulin CA: Lethal intracranial complications following inflation in the external auditory canal in treatment of serious otitis media and due to defects in the petrous bone. Aeta Otolaryngol (Stoekh) 60:407-421, 1965 2. Bell WE, Joynt R J, Sahs AL" Low spinal fluid pressure syndromes. Neurology 10:512-514, 1960 3. Chiari H: Ober einen Fall yon Luftansammlung in den Ventrikeln des menschlichen Gehirns. Z Heilkd 5:383-390, 1884 4. Dandy WE: Treatment of rhinorrhea and otorrhea. Arch Surg 49: 75-85, 1944 5. Finsnes KA: Lethal intracranial complication following air insufflation with a pneumatic otoscope. Acta Otolaryngol (Stock) 75: 436-483, 1973 6. Kessler LA, Stern WZ: Ventriculo-pleural shunt for hydrocephalus. Case report of an unusual complication. J Pediatr 60:418-420, 1962 7. Luckett WH: Air in the ventricles of the brain, following a fracture of the skull. Surg Gyneeol Obstet 17:237-240, 1913 8. Markham JW: The clinical features of pneumocephalus based upon a survey of 284 cases with report of 11 additional cases. Acta Neurochirurgica 16:1-78, 1967 9. Ray BS, Bergland RM: CSF fistula: clinical aspects, techniques of localization and methods of closure. J Neurosurg 30:399-405, 1969 10. Schaltenbrand G: Die akute aliquorrhoe. Verh Dtsch Ges Inn Med 52:473-481, 1940
This work was supported in part by USPHS Training Grant NINDS 5593. This paper was presented in part at the Annual Meeting of the Neurosurgical Society of America, 1973, Bermuda. Address reprint requests to: Charles B. Wilson, M.D., Department of Neurological Surgery, University of California School of Medicine, San Francisco, California 94143.
633
