U92:ZPENE903XA
BAHR
Case Report Pediatr Neurosurg 1998;28:35–41
Keyvan Abtin B. Gregory Thompson Marion L. Walker Department of Neurological Surgery, Division of Pediatric Neurosurgery, University of Utah, Primary Children’s Medical Center, Salt Lake City, Utah, USA
Received: February 27, 1998 Accepted: March 3, 1998
Basilar Artery Perforation as a Complication of Endoscopic Third Ventriculostomy
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Key Words Third ventriculostomy Neuroendoscopy Pseudoaneurysm Complication Hydrocephalus Hemorrhage
Abstract The morbidity and mortality associated with third ventriculostomy has decreased significantly over the past 75 years since its introduction by Walter Dandy. Now more commonly performed using an endoscopic method, the significant morbidity of third ventriculostomy has dropped to approximately 5%; essentially that associated with ventriculoscopy in general. However, the possible complication of massive subarachnoid hemorrhage resulting from perforation of the basilar artery or its branches in the course of fenestration of the floor of the third ventricle has only recently been reported. In our case, subsequent to a vascular injury, a pseudoaneurysm developed at the site of vascular perforation, which was then appropriately controlled. The patient has since made a full recovery. Our goal is to remind the endoscopist of this unusual complication and to discuss our management strategies. OOOOOOOOOOOOOOOOOOOOOO
Introduction Third ventriculostomy has come a long way since its introduction in 1922 by Dandy [1]. Initially the procedure was performed as an open craniotomy, where the lamina terminalis was fenestrated through a subfrontal approach [2]. One year later, Mixter [3] reported the first successful case of an endoscopic third ventriculostomy using an urethroscope. Since then, there have been a number of methods introduced including open, percutaneous, stereotactic, endoscopic or modifications of the above [4–19]. Currently most commonly performed using an endoscopic
This paper was presented at the ASPN meeting, Lana’i, Hawaii, January 1998.
ABC Fax + 41 61 306 12 34 E-Mail
[email protected] www.karger.com
© 1998 S. Karger AG, Basel 1016–2291/98/0281–0035$15.00/0 This article is also accessible online at: http://BioMedNet.com/karger
technique, the success rate of this procedure has been reported to be greater than 60% in selected patients presenting with late onset obstructive hydrocephalus [16, 19– 23]. The morbidity and mortality associated with third ventriculostomy has improved along with improved technology and the refinement of the operative technique. Mortalities as high as 33% were reported in earlier publications with open fenestration of the lamina terminalis [24]. The morbidity and mortality rate subsequently decreased as the surgeon became more experienced and as the degree of invasiveness decreased. Currently, the procedure is performed endoscopically in many centers with no reported mortalities and essentially a morbidity equal to that associated with ventriculoscopy in general [4, 25, 26]. Teo et al. [26] published their complication rate with
Marion L. Walker, MD Division of Pediatric Neurosurgery Primary Children’s Medical Center 100 N. Medical Dr., Suite 2400 Salt Lake City, UT 84113 (USA)
U92:ZPENE903XA
BAHR
tis. All of these patients made a full recovery. One mortality occurred 10 months postoperatively in a child who herniated acutely after a 2-week course of increasing headaches, vomiting, and lethargy suggestive of progressive hydrocephalus, presumably due to delayed failure of the third ventriculostomy. The patient was not brought to medical attention until she was deeply unconscious. She died during transport to our facility. Thus our overall complication rate reached 6% in this series. Case Report
Fig. 1. Axial MR T2-weighted images showing enlarged lateral and third ventricles with a slightly enlarged fourth ventricle. There is no sign of transventricular flow.
173 consecutive endoscopic procedures. Significant morbidity was seen in 7% of the cases, whereas another 13% of the procedures were complicated with an ‘intraoperative problem’, the incidence of which decreased significantly with the surgeon’s experience. Most of the significant complications associated with endoscopic third ventriculostomy reported in Teo’s series were related to hypothalamic-pituitary axis dysfunction. At the Primary Children’s Medical Center at the University of Utah, we performed 71 endoscopic third ventriculostomies on 97 patients between June 1992 and December 1996. We had to abandon the procedure in the remaining 26 patients due to difficulties encountered intraoperatively, such as unfavorable anatomy. Significant morbidity included 1 transient decrease in level of consciousness, 1 transient herniation syndrome, 1 basilar artery perforation, and 2 postoperative cases of ventriculi-
36
Pediatr Neurosurg 1998;28:35–41
An 18-year-old female, diagnosed with compensated hydrocephalus at age 2 weeks, had been followed by the senior author and monitored closely over the years. She remained shunt-free. She had repeated head computed tomography (CT) scans, neuropsychiatric evaluations, and visual examinations. Her brain magnetic resonance imaging (MRI) revealed dilated lateral and third ventricles, a normal fourth ventricle and an absence of other intracranial pathology (fig. 1). She was found to have signs of chronic mild papilledema on a routine eye examination. This finding was absent on a prior visual examination the year before. The findings were corroborated by 2 pediatric ophthalmologists. She had essentially unchanged ventricular size on CT scan. Following admission, she underwent 48 h of intracranial pressure (ICP) monitoring. Her ICP remained in the normal range (5–12 mm Hg) except for sustained elevations up to 30–40 mm Hg range during short-lived (15–20 s) nocturnal bradypneic episodes associated with mild oxygen desaturation and hypercapnia. She remained headache-free. Following this hospital course and a discussion with the patient and her family, endoscopic third ventriculostomy was proposed in an attempt to avoid a ventriculoperitoneal shunt. It was recognized that she was not the ideal candidate for endoscopic third ventriculostomy because she did not have complete aqueductal stenosis. However, given the disproportionately larger lateral and third ventricles to fourth ventricular size, and the relative success of this operation in older patients, she was felt to be a reasonable operative candidate. An endoscopic third ventriculostomy was performed through a 7-mm coronal craniostomy using a 1.2-mm NeuroNavigational neuroendoscope. The right lateral ventricle was cannulated using a 7-fr peelaway sheath. The foramen of Monro was easily identified and normal anatomy of the third ventricle was noted with a thin, translucent floor. The tip of the endoscope was used in an attempt to fenestrate the floor of the third ventricle in the midline anterior to the mamillary bodies and posterior to the infundibular recess. The floor was found to be rubbery with relatively high elastic compliance. At this time we withdrew the endoscope to the tip of our peelaway sheath and under direct visualization through the endoscope, in order to avoid trauma to the structures at the foramen of Monro, the sheath was advanced into the third ventricle. With the peelaway sheath in position in the upper portion of the third ventricle, another instrument, if needed, could be passed alongside the endoscope. The fenestration was reattempted this time by pressing the floor against the dorsum of the sella turcica. Upon penetrating the third ventricular floor, the field became full of blood. The endoscope was removed, following which bright red blood under significant arterial pressure exited the sheath. The surgical team was informed of the possible
Abtin/Thompson/Walker
U92:ZPENE903XA
BAHR
a
b
Fig. 2. Postoperative head CT scan showing diffuse subarachnoid blood in the cisterns (a) and intraventricular hemorrhage (b).
implications, non-cross-matched blood was obtained while crossmatched blood was in preparation. Following the removal of the endoscope, an external ventricular drain was placed through the sheath into the third ventricle and we proceeded with copious amount of irrigation with lactated Ringer’s solution. The irrigation was continued for approximately 20 min while the patient received replacement blood for a total of 350 ml of blood loss. Her vital signs remained stable throughout the operation with systolic blood pressures remaining above 90 mm Hg. After 20 min of continuous third ventricular irrigation the cerebrospinal fluid (CSF) cleared, allowing endoscopic visualization. The third ventricular floor revealed a fenestration and a small clot remained in the third ventricle. We did not attempt to pass the endoscope through the fenestration. Approximately 5 ml of thrombin was infused through the external ventricular drain directly into the third ventricle. The drain was left in the frontal horn of the right lateral ventricle. An immediate head CT scan was obtained while the patient remained intubated and under anesthesia. The scan revealed diffuse subarachnoid hemorrhage (SAH) and blood in the lateral ventricles (fig. 2). The patient awoke immediately following the surgery without any focal neurological deficits and was extubated in the intensive care unit. Slow CSF drainage was allowed over the next several days. On the 3rd postoperative day, a four-vessel cerebral angiogram demonstrated a 3-mm pseudoaneurysm on the P1 segment of the right posterior cerebral artery (PCA) (fig. 3). She subsequently underwent a right pterional craniotomy during which the pseudoaneurysm was directly clipped at its presumed neck. Intraoperatively, a through-and-through penetration of the P1 segment of the right PCA was discovered. She fully recovered from mild basilar artery vasospasm but had no other problems. The immediate postoperative angiogram showed successful clipping of the pseudoaneurysm while preserving the P1 segment. However, on her
1-month repeat angiogram, the pseudoaneurysm had returned, displacing the clip (fig. 4). Having demonstrated on angiography that she had a persistent fetal origin of the right posterior cerebral artery with an atretic P1 (fig. 5), she underwent a repeat craniotomy with successful trapping of the right P1 segment without compromise to any perforating vessels. The immediate postoperative angiogram showed successful obliteration of the pseudoaneurysm (fig. 6). This was confirmed by 1-month and 1-year follow-up angiograms. Because of persistent hydrocephalus she underwent ventriculoperitoneal shunt placement following recovery from the first craniotomy. She has made excellent recovery from her surgeries. She remains intact with good cognitive function, normal visual acuity and holds a full-time job following graduation from high school.
Basilar Artery Perforation as a Complication of Endoscopic Third Ventriculostomy
Pediatr Neurosurg 1998;28:35–41
Discussion Significant advances in neuroendoscopic technology have made third ventriculostomy a relatively safe operation for the treatment of noncommunicating hydrocephalus in selected patients. The morbidity and mortality associated with endoscopic procedures, however, can be significantly reduced as the surgeon gains experience [26]. Yet it is apparent that this procedure is not completely risk-free. There remains a 7% significant morbidity risk in all neuroendoscopic procedures, including third ventriculostomy. Our own experience has shown a 4% significant morbidity, usually temporary, and 1 patient mortality (1%), which was not a direct operative complication.
37
U92:ZPENE903XA
BAHR
Fig. 3. Cerebral angiogram taken on the 3rd postoperative day following endoscopic third ventriculostomy. AP view of the right vertebral artery injection shows a pseudoaneurysm on the P1 segment of the right posterior cerebral artery.
Fig. 4. Cerebral angiogram one month postoperative from the clipping of the pseudoaneurysm reveals recurrence of the pseudoaneurysm on this AP view of the right vertebral artery injection.
This case represents a rare complication only recently reported in the medical literature [27]. Although to our knowledge several similar cases in North America have occurred, the results and patient outcomes were not as fortunate as in our case. We feel that forces other than just good fortune played a role in our patients good outcome.
38
Pediatr Neurosurg 1998;28:35–41
Intraoperative placement of the peelaway sheath in the third ventricle played by far the most important role in her survival. It allowed decompression of the premesencephalic cistern and the third ventricle from the rapidly accumulating hematoma and hence prevented an immediate herniation syndrome. The sheath also provided us
Abtin/Thompson/Walker
U92:ZPENE903XA
BAHR
Fig. 5. Lateral view of the right internal cerebral artery injection shows the persistent fetal origin of the right posterior cerebral artery.
Fig. 6. An AP view of the right vertebral artery injection 1 week postoperative from the trapping of the right P1 segment. There is no filling of the right posterior cerebral artery and this produced ultimate control of the pseudoaneurysm. This was confirmed on an angiogram obtained on 1-year follow-up.
with the means to quickly pass an external ventricular drain into the third ventricle, allowing irrigation with lactated Ringer’s solution, thus assisting with clot removal and in obtaining hemostasis. When the floor of the third ventricle was found to be very patulous, not tense as is the usual circumstance, it
would have been wise to try to perform the fenestration by some other means than penetration with a semiblunt object, i.e., the 1.2-mm endoscope. A sharp pointed laser fiber (without laser energy) may have worked well. Grasping forceps, to hold the floor while penetration is accomplished, is another option. We strongly advise against the
Basilar Artery Perforation as a Complication of Endoscopic Third Ventriculostomy
Pediatr Neurosurg 1998;28:35–41
39
U92:ZPENE903XA
BAHR
use of lasers or any type of hot, coagulating energy. The vessels of the circle of Willis are immediately beneath the third ventricle and routine use of these devices will eventually result in injury to a major vessel or to small perforators to the brain stem [19]. The recent report of a pseudoaneurysm following endoscopic third ventriculostomy by McLaughlin et al. [27] appears to have been caused by penetration of the floor of the third ventricle by laser energy. Postoperatively, our patient was treated as any fresh SAH patient would be from a ruptured cerebral aneurysm. With the presumption that a major vessel of the circle of Willis had been perforated, a cerebral angiogram was obtained in order to assess the extent of the vascular injury. The angiogram was delayed until the 3rd postoperative day in order to provide time for genesis of a possible pseudoaneurysm and hence to reduce a possible false-negative result. Yilmaz et al. [28] performed a retrospective analysis of their data on delayed arterial injury and found that the initial exploration or arteriogram missed the resulting vascular injury in the majority of those patients who subsequently presented with pseudoaneurysm (52.5%), arteriovenous fistula (35%), or occlusion (12.5%). If the initial angiogram fails to show any signs of vascular injury, either because of vasospasm or ‘cut off’ of a vessel, a repeat angiogram is warranted to further elucidate a possible vascular abnormality [29]. If the surgeon strongly suspects a major vascular injury, and if the initial angiogram is negative for a vascular anomaly, even after 3 days, then a repeat angiogram is indicated at 7–10 days after the initial insult. Finally, once the diagnosis of pseudoaneurysm is established, one should make a prompt referral to a cerebrovascular surgeon for timely control of the pseudoaneurysm. The recently published case report by McLaughlin et al. [27] of a case similar to ours appears to be secondary to the use of laser energy during endoscopic third ventricu-
lostomy. A cerebral angiogram 3 days following the initial injury was negative. The pseudoaneurysm was discovered a month later after acute rupture and a massive subarachnoid hemorrhage. These authors have since changed their technique and use blunt instruments to perforate the floor of the third ventricle. In the past, we have advocated against the use of lasers or any type of thermal injury to perform this procedure [19]. Furthermore, a repeat angiogram is warranted in cases where the initial angiogram failed to demonstrate a vascular injury [29]. We do not feel that performing a craniotomy immediately after the vessel perforation occurred would have been the right choice in the management of the hemorrhage or the vascular injury. Retrospectively, a craniotomy would have been more time-consuming and would not have allowed for timely decompression of the rapidly filling cisterns and ventricles and would have resulted in perhaps more serious complications or even her death on the operating table.
Conclusions Serious vascular injury during a neuroendoscopic procedure such as third ventriculostomy is an underreported complication with devastating implications. Immediate intraoperative management plays an essential role in patient survival. In our case, placement of the sheath in the third ventricle was key, followed by irrigation of the ventricles with lactated Ringer’s solution. Finally, a cerebral angiogram is indicated in order to assess the injury. Of course, one should remember that the best management strategy would be to abandon the procedure of third ventriculostomy in those where difficult anatomy, such as a rubbery patulous floor of the ventricle, is encountered. In such a case, simply placing a shunt would address the problem, as it ultimately did in our case.
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
References
1 Dandy WE: Cerebral ventriculoscopy. Bull Johns Hopkins Hosp 1922;33:189. 2 Dandy WE: An operative procedure for hydrocephalus. Bull Johns Hopkins Hosp 1922;33: 189–190. 3 Mixter WJ: Ventriculoscopy and puncture of the floor of the third ventricle. Boston Med Surg J 1923;188:277–278. 4 Kelly P, Goerss S, Kall B, Kispert D: Computed tomography based stereotactic third ventriculostomy. Neurosurgery 1986;18:791–794.
40
5 Kelly PJ: Stereotactic third ventriculostomy in patients with nontumoral adolescent/adult onset aqueductal stenosis and symptomatic hydrocephalus. J Neurosurg 1991;75:865–873. 6 Manwaring K, Rekate H, Kaplan A: Hydrocephalus management by endoscopy (abstract). Ann Neurol 1989;26:488. 7 McNickle H: The surgical treatment of hydrocephalus. A simple method of performing third ventriculostomy. Br J Surg 1947;334:302–307.
Pediatr Neurosurg 1998;28:35–41
8 Morello G, Migliavacca F: Third ventriculostomy (Stookey and Scarff operation) in the treatment of benign aqueductal stenosis. Acta Neurochir 1959;7:417–424. 9 Musolino A, Soria V, Munari C, Devaux B, Merienne L, Constans JP, Chodkiewicz JP: Stereotaxic ventriculocisternostomy in the treatment of triventricular obstructive hydrocephalus. Apropos of 23 cases. Neurochirurgie 1988;34:361–373.
Abtin/Thompson/Walker
U92:ZPENE903XA
BAHR
10 Paine K, McKissock W: Aqueduct stenosis. Clinical aspects, and results of treatment by ventriculocisternostomy (Torkildsen’s operation). J Neurosurg 1955;12:127–145. 11 Patterson RH Jr, Bergland RM: The selection of patients for third ventriculostomy bases on experience with 33 operations. J Neurosurg 1968;29:252–254. 12 Pierre-Kahn A, Renier D, Bombois B, Askienay S, Moreau R, Hirsch JF: Role of the ventriculocisternostomy in the treatment of noncommunicating hydrocephalus. Neurochirurgie 1975;21:557–569. 13 Poblete M, Zamboni R: Stereotaxic third ventriculocisternostomy. Confin Neurol 1975;37: 150–155. 14 Putnam T: Treatment of hydrocephalus by endoscopic coagulation of the choroid plexus. Description of a new instrument and preliminary report of results. N Engl J Med 1934;210:1373– 1376. 15 Putnam TJ: The surgical treatment of infantile hydrocephalus. Surg Gynecol Obstet 1942;76: 171–182. 16 Sainte-Rose C: Third ventriculostomy; in Manwaring KH, Crone K (eds): Neuroendoscopy, New York, Liebert, 1992, pp 47–62.
Basilar Artery Perforation as a Complication of Endoscopic Third Ventriculostomy
17 Sayers MP, Kosnik EJ: Percutaneous third ventriculostomy: Experience and technique. Childs Brain 1976;2:24–30. 18 Scarff JE: Third ventriculostomy by puncture of the lamina terminalis and the floor of third ventricle. J Neurosurg 1966;24:935–943. 19 Walker M, Petronio J, Carey C: Ventriculoscopy; in Cheek W (ed): Pediatric Neurosurgery: Surgery of the Developing Nervous System. Philadelphia, Saunders, 1994, pp 572–581. 20 Drake JM: Ventriculostomy for treatment of hydrocephalus. Neurosurg Clin N Am 1993;4: 657–666. 21 Jones RF, Kwok BC, Stening WA, Vonau M: Neuroendoscopic third ventriculostomy. A practical alternative to extracranial shunts in non-communicating hydrocephalus. Acta Neurochir Suppl (Wien) 1994;61:79–83. 22 Jones RF, Kwok BC, Stening WA, Vonau M: Third ventriculostomy for hydrocephalus associated with spinal dysraphism: indications and contraindications. Eur J Pediatr Surg 1996;6 (suppl 1):5–6.
23 Teo C, Jones R: Management of hydrocephalus by endoscopic third ventriculostomy in patients with myelomeningocele (discussion). Pediatr Neurosurg 1996;25:57–63. 24 Wetheimer P, Mansuy L: L’incision de la lame sus-optique: Procédé de traitement de l’hydrocéphalie interne par drainage du troisième ventricle. J Chir (Paris) 1938;51:838–848. 25 Jones RFC, Teo C, Stening WA, Kwok BCT: Neuroendoscopic third ventriculostomy; in Manwaring KH, Crone K (eds): Neuroendoscopy, New York, Liebert, 1992, pp 63–77. 26 Teo C, Rahman S, Boop FA, Cherny B: Complications of endoscopic neurosurgery (discussion). Childs Nerv Syst 1996;12:248–253. 27 McLaughlin M, Wahlig J, Kaufmann A, Albright A: Traumatic basilar aneurysm after endoscopic third ventriculostomy: Case report. Neurosurgery 1997;41:1400–1404. 28 Yilmaz A, Arslan M, Demirkilic U, Ozal E, Kuralay E, Tatar H, Ozturk O: Missed arterial injuries in military patients. Am J Surg 1997; 173:110–114. 29 du Trevou M, van Dellen J: Penetrating stab wounds to the brain: The timing of angiography in patients presenting with the weapon already removed. Neurosurgery 1992;31:905– 911.
Pediatr Neurosurg 1998;28:35–41
41
