High Intracranial Pressure, Brain Herniation and Death in Cerebral Venous Thrombosis
To the Editor:
Sir, the study by Canhao and colleagues made the important point that the most frequent cause of death in patients with cerebral venous thrombosis (CVT) is transtentorial herniation and that these patients may potentially have benefited from decompressive hemicraniectomy.1 We would like to corroborate this argument by providing evidence for a rise in intracranial pressure (ICP) preceding brain herniation and death in a patient with CVT.
A 29-year-old pregnant woman who presented with confusion and vomiting was admitted to a district general hospital. A right-sided weakness developed within 2 days. A CT brain scan showed a left temporal hemorrhage without mass effect. Her Glasgow Coma Scale score dropped to 8/15; she was intubated and transferred to the neurocritical care unit at the National Hospital. A repeat CT demonstrated worsening of the hemorrhage with obliteration of the 3rd and 4th ventricles, and the diagnosis of a CVT was confirmed by magnetic resonance venography. At this point, the left pupil became fixed. An ICP bolt was inserted (opening pressure 50 mm Hg; Figure). ICP targeted management (propofol, fentanyl, midazolam, ventilation to maintain a pCO2 4.0 to 4.5 kPa and IV norepinephrine to main blood pressure) and anticoagulation with IV heparin were started. Because an ICP <20 mm Hg could not be maintained, paralysis and moderate hypothermia were initiated. Despite these measures ICP continued to rise. A treatment trial with thiopental to lower ICP failed. Both pupils became fixed and dilated on day 6 on ICU. Several intractable peaks of ICP (>60 mm Hg) were followed by development of diabetes insidipus necessitating treatment with desmopressin. At 8:00 am on the 7th day, periods of ventricular tachycardia and fibrillation started to appear leading to severe hemodynamic compromise and elevated ICP. The clinical diagnosis of brain stem herniation was made and treatment was demescalated after an informed discussion with the family took place.
Mortality in CVT has decreased over the last decades from 30% to 50% to ≈4.3% in the acute phase.1,2,3 There are occasional reports of decompressive hemicraniectomy performed successfully in patients in whom medical treatment failed.4 It is of note that all 3 patients in this report already showed signs of brain herniation at time of operation. The authors pointed out that indications for surgical intervention are almost completely lacking. The decision to proceed with surgery implies that treatment (heparin) needs to be discontinued. In an individual case this may be a difficult decision because of the arguably beneficial effect (small sample sizes and large confidence intervals).5 Furthermore, there is no guide toward the best timing for surgical intervention. Fixed and dilated pupils may be too late a sign and repeated brain imaging is logistically difficult in the critically ill patient. Continuous ICP and arterial blood pressure monitoring as performed in the present case provides important data at the bedside. There is a need to investigate whether decompressive hemicraniectomy would be of benefit in those patients in whom medical management of ICP fails.
Canhao P, Ferro JM, Lindgren AG, Bousser MG, Stam J, Barinagarrementeria F; ISCVT Investigators. Causes and predictors of death in cerebral venous thrombosis. Stroke. 2005; 36: 1720–1725.
Stam J, de Bruijn, SFTM, deVeber G. Anticoagulation for cerebral sinus thrombosis (Review). The Cochrane Database of Systematic Reviews. 2001; 4: CD002005.
We thank Petzold et al for illustrating with this case report the results of our study, regarding the causes of death in patients with cerebral vein and sinus thrombosis.1 Probably, the patient here presented could have been saved if decompressive craniectomy had been performed before fatal clinical deterioration occurred.
We consider that the next step for evaluating the role of this intervention should be to carry out an international registry including cerebral vein and sinus thrombosis patients submitted to decompressive craniectomy.