Hemorrhagic complications of external ventricular drainage.

OBJECTIVE: Despite the widespread use of external ventricular drainage (EVD), the frequency of associated hemorrhagic complications remains unclear. This retrospective study examined the frequency of hemorrhagic complications of EVD and attempted to discern associated risk factors. METHODS: Treatment records from 160 patients admitted during a 2.5-year period who required EVD placement were reviewed. Indications for placement of EVD included acute complications of cerebrovascular disease (n = 94), traumatic brain injury (n = 36), primary hydrocephalus (n = 16), and tumor (n = 14). Patients received either a 3.0 or 2.5-mm outer diameter ventricular catheter (n = 82 and 78, respectively). Postinsertion computed tomographic scans were obtained within 24 hours on all patients and were analyzed for any new hemorrhage related to the ventricular catheter. Patient age, sex, catheter type, and dimensions of hemorrhage were also analyzed. RESULTS: The incidence of EVD-related hemorrhage was 33 +/- 0.04%. However, the incidence of detectable change in the clinical neurological examination was 2.5%. A significant proportion of EVD-related hemorrhages were small (<4 cm), punctate, intraparenchymal hematomas. Patients with cerebrovascular disease exhibited an increased incidence (39%) of hemorrhage. The mean volume of intraparenchymal hemorrhage was larger in patients who received the 2.5-mm ventricular catheter, as well as those admitted for cerebrovascular disease. CONCLUSION: Hemorrhagic complications of EVD placement are more common than previously suspected. Admitting diagnosis seems to have an effect on the development of an associated hemorrhage and its size. Catheter gauge has an effect on hematoma volume. Most of the hemorrhages seen on postinsertion computed tomographic scans do not cause detectable changes in the clinical examination.

A prospective evaluation of the value of repeat cranial computed tomography in patients with minimal head injury and an intracranial bleed.

BACKGROUND: Patients with minimal head injury (MHI) and intracranial bleed (ICB) detected on cranial computed tomography (CT) scan routinely undergo a repeat cranial CT within 24 hours after injury to assess for progression of intracranial injuries. While this is clearly beneficial in patients with a deteriorating neurologic status, it is of questionable value in patients with a normal neurologic examination. The goal of this study was to prospectively assess the value of a repeat cranial CT in patients with a MHI and an ICB who have a normal neurologic examination. METHODS: A prospective analysis of all adult patients admitted to a Level I trauma center after blunt trauma causing a MHI (defined as the loss of consciousness or posttraumatic amnesia with a Glasgow Coma Scale (GCS) score of greater or equal to 13) and an ICB on the initial cranial CT during a 12-month period (July 2002 through July 2003) was performed. All patients with MHI were prospectively evaluated and followed until discharge. Data collected included demographics, neurologic examination and findings on the initial and repeat cranial CT scan. Outcome data included neurologic deterioration, neurosurgical intervention, and Glasgow Outcome Scale (GOS) on discharge. RESULTS: In all, 161 consecutive patients with MHI and a positive cranial CT scan were identified. The initial cranial CT lead to a neurosurgical intervention (1 craniotomy, 4 intracranial pressure monitors) in 4% of cases. The remaining 130 patients who met inclusion criteria, underwent a repeat cranial CT scan within 24 hours postadmission. Ninety nine (76%) patients had a normal neurologic examination at the time of their repeat cranial CT. After the repeat cranial CT none required immediate neurosurgical intervention or had delayed neurologic deterioration related to their head injury. Fifteen patients underwent additional neuroradiologic studies but none showed further progression of their ICB or lead to a change in management. One patient died from non-traumatic brain injury related causes and of the remaining 26 patients, 98% had an overall favorable GOS score (> 3) on discharge. In this group of patients with MHI and ICB, the negative predictive value of a normal neurologic examination was 100%. CONCLUSIONS: Repeat cranial CT, in patients with a MHI and a normal neurologic examination, resulted in no change in management or neurosurgical intervention and is therefore not indicated. A multicenter prospective study would further validate these conclusions, reduce unnecessary CT scans, and likely improve our current standard of care in these patients.