HAGEN B. HUTTNER, M.D., MARTIN KÖHRMANN, M.D., CHRISTIAN BERGER, M.D., DIMITRIOS GEORGIADIS, M.D., AND STEFAN SCHWAB, M.D.

Transcription

1 J Neurosurg 105: , 2006 Influence of intraventricular hemorrhage and occlusive hydrocephalus on the long-term outcome of treated patients with basal ganglia hemorrhage: a case control study HAGEN B. HUTTNER, M.D., MARTIN KÖHRMANN, M.D., CHRISTIAN BERGER, M.D., DIMITRIOS GEORGIADIS, M.D., AND STEFAN SCHWAB, M.D. Department of Neurology, University of Heidelberg, Germany Object. Only limited data exist addressing the long-term outcome of patients with ganglionic hemorrhage complicated by intraventricular hemorrhage (IVH) and hydrocephalus that requires an external ventricular drain (EVD). The aim of this study was to compare the long-term outcome of patients with pure ganglionic hemorrhage and those requiring an EVD due to additional IVH and hydrocephalus. Methods. For this study, 116 patients with supratentorial ganglionic hemorrhage and occlusive hydrocephalus were screened. To avoid any bias the authors excluded all patients with nonprimary intracerebral hemorrhage as well as those who received no treatment. Forty patients with IVH and subsequent hydrocephalus were examined, and 40 more patients with pure ganglionic hematomas were matched with regard to age, sex, Glasgow Coma Scale score, need for mechanical ventilation, and, in particular, hematoma volume. Outcome analysis was performed using the Barthel Index and the modified Rankin Scale (mrs). The mean hematoma volume was ml. The overall long-term outcome was unfavorable (mrs Score 4 6) in 59% of all patients. Good outcome (mrs score 4) was observed in 25 (62.5%) of 40 patients with hematoma volumes less than 25 ml, compared with eight (20%) of 40 who had hematoma volumes greater than 25 ml (p 0.05, chi-square test). The need for an EVD was not associated with a worse long-term outcome in patients with comparable hematoma volumes. In contrast, the duration of treatment in the intensive care unit was longer for patients with EVDs than for those who had pure ganglionic hematomas (16 [range 5 29] days compared with 8 [range 2 19] days; p 0.05, Mann Whitney U-test), regardless of hematoma volume. Conclusions. The long-term outcome of treated patients with supratentorial ganglionic hemorrhage with ventricular involvement and occlusive hydrocephalus is comparable to that of patients with similar hematoma volumes but no IVH. KEY WORDS intraventricular hemorrhage hydrocephalus external ventricular drain outcome I NTRACEREBRAL hemorrhage accounts for 10 to 15% of all patients who suffer strokes. 4,9 Morbidity and mortality rates in patients with ICH are higher than in those with ischemic stroke. 25 As a general rule, one third of patients with ICH are discharged from the hospital with no or only minor deficits, one third have a severe disability, and one third die immediately or during the hospital stay. The 1-year mortality rate for patients with ICH ranges between 37 and 47%. 6 Prognostic factors for poor outcome are as follows: 1) an initially reduced level of consciousness; 2) a large hematoma size; 3) a patient s age; and 4) the presence of IVH. 13,16,23,27 Abbreviations used in this paper: BI = Barthel Index; CSF = cerebrospinal fluid; CT = computed tomography; EVD = external ventricular drain; GCS = Glasgow Coma Scale; ICH = intracerebral hemorrhage; ICP = intracranial pressure; ICU = intensive care unit; IVH = intraventricular hemorrhage; MR = magnetic resonance; mrs = modified Rankin Scale; SD = standard deviation. Intraventricular hemorrhage occurs in 3 to 10% of all patients with intracerebral hematomas 5 and can be classified as primary and secondary IVH. Primary IVH originates from intraventricular sources, whereas secondary IVH, which predominates in 70% of cases, reflects an extension of intraparenchymal hemorrhage into the ventricular system. 2,8,28 The use of an EVD to divert CSF and blood can result in a reversal of obtundation. Although ventriculostomy is effective in reducing elevated ICP, morbidity cannot be adequately addressed with an EVD alone. 7,21 Despite obvious improvement in acute stroke care in the last few years, IVH remains associated with a high case fatality rate and poor functional outcome, regardless of the treatment options used. Additionally, it leads to long-term hospital stays and consequently to a high cost of care. In various studies it has been reported that the presence of IVH and of occlusive hydrocephalus are independent risk factors for an increased 30-day mortality rate in patients with ICH. 7,19,24,28 The long-term outcome of patients presenting with these features, however, has not been studied in de- 412 J. Neurosurg. / Volume 105 / September, 2006

2 Influence of IVH and occlusive hydrocephalus on outcome of ICH tail. Our objective was to investigate the long-term outcome in patients with supratentorial ganglionic hematoma with or without IVH plus hydrocephalus. To study possible differences in the long-term outcome after 12 months, we analyzed patient pairs from our institutional database. To avoid any bias we excluded all patients who received no treatment (those with do not resuscitate orders). The remaining patients were matched for hematoma volume and location, age, sex, GCS score, and need for mechanical ventilation, and were classified as presenting with either pure ganglionic hemorrhage or additional IVH and hydrocephalus requiring EVD placement. Patient Selection Clinical Material and Methods Data for this study were obtained from a prospectively organized database of patients treated in our department. The following steps were undertaken: first, of 1372 patients with ICH, 412 with primary hypertensive ganglionic hemorrhage who did not undergo surgical evacuation of the hematoma were included. Patients with warfarin-related ICH were excluded, as were those with sub- and epidural hematomas, subarachnoid hemorrhage, and ICH caused by trauma or tumors. Next, of these 412 patients, 116 with occlusive hydrocephalus were identified in a second step, and 76 who received an EVD were identified in a third step. An EVD was not placed in 40 patients with unfavorable prognoses and do not resuscitate orders, because of signs of herniation on admission in combination with severe comorbidity, age older than 80 years, or lack of relatives consent for neurocritical care. Four patients treated with EVDs suffered from cardiopulmonary complications, and do not resuscitate orders were decided on during the course of treatment. These patients were not enrolled. We next excluded nine patients in whom an EVD was inserted due to mass effect with midline shift. Our aim was to isolate patients who received EVDs due to ventricular involvement with subsequent obstructive hydrocephalus (63 patients). Finally, patients who received intraventricular fibrinolytic therapy for clot lysis were not considered (23 individuals). Thus, 40 patients remained for further analysis. To design a case control study, we subsequently selected patients from our database who had a ganglionic hematoma but no IVH and no or only a slight mass effect that required no EVD (138 patients). The matched pairs of patients were chosen from those who had IVH and EVD, and with regard to the following parameters, which were prioritized (the most important parameter for matching was the hematoma volume). When volumes were similar, the location of the thalamic hematoma, followed by age ( 3 years), then sex, then GCS score on admission ( 1 point) were used for matching. Eight patients remained who were not definitively matched. For them we chose the duration of ventilation as well as serum glucose levels on admission for definitive matching. We determined the clinical parameters by reviewing the medical records and the neuroimaging parameters by reviewing the cerebral CT scans and MR images. Neuroimaging Data J. Neurosurg. / Volume 105 / September, 2006 The ICH was diagnosed immediately after hospital admission by analyzing cranial CT scanning (Somatom Volume Zoom; Siemens, Erlangen, Germany) or MR imaging studies (Symphony [1.5-tesla magnet]; Siemens). Ganglionic hematomas were defined to include thalamic lesions. In cases of ventricular involvement, the site of the ventricles affected was noted using the Graeb score. 11 Hydrocephalus was determined by enlargement of the lateral ventricles on cerebral CT scans on admission or on follow-up CT scans obtained 12 hours later, which were obtained independently of those acquired to evaluate possible neurological worsening. Hydrocephalus was evaluated by measuring the bicaudate index (it was considered present when it exceeded the 95th percentile for age) and the mean temporal horn diameter. 29 The hematoma volume from ICH was calculated using the initial cerebral CT or MR imaging study according to the formula for ellipsoids (A B C/2, with A, B, and C representing the radii in all three dimensions), which has been demonstrated to estimate hematoma volume reliably. 10,15 The portion of blood in the intraventricular region was not considered in the hematoma volume measurement. Hematoma volumes are given in milliliters. Clinical Management All patients received standard medical treatment according to an institutional protocol for ICH, including early securing of the airway by endotracheal intubation at GCS scores of 7 or less, and sedation with fentanyl and midazolam. Blood pressure was monitored continuously by using intraarterial catheters and was treated when systolic pressure exceeded 185 mm Hg or diastolic pressure exceeded 125 mm Hg. In patients with IVH, an EVD (Spiegelberg III; Spiegelberg GmbH & Co., Hamburg, Germany) was inserted as soon as occlusive hydrocephalus was diagnosed. In patients with purely ganglionic hemorrhage, an ICP probe (Spiegelberg I; Spiegelberg GmbH & Co.) was placed ipsilateral and rostral to the hematoma and monitoring was only initiated in sedated patients who were receiving ventilation. The ICP monitoring was performed continuously by using an intraabdominal pressure monitor (IAP 32.1; Spiegelberg GmbH & Co.). A rise in ICP was defined as pressure levels of 20 mm Hg or more for longer than 5 minutes, and was treated with administration of mannitol according to the extent of the increase (100 ml 4 /day for a slight increase and 200 ml 6 /day for a large increase) and HyperHAES (250 ml 2 4 /day according to the extent of the increase). The ICP probe was removed when pressure was at or below 20 mm Hg for more than 24 hours or in cases of EVD infection. In patients with EVDs the CSF was analyzed every 2nd day for protein and the cell count. The EVD was replaced according to established criteria in cases of infection or dislocation as well as in cases of insufficient drainage due to obstruction. Intraventricular administration of fibrinolytic or antibiotic agents was not performed. The EVD was clamped if ICP was 20 mm Hg or less for more than 24 hours, and it was removed after another 24 hours if control cerebral CT scans demonstrated no evidence of enlarging ventricles. The EVD was changed routinely after 7 days. Ventriculoperitoneal shunt placement was indicated when occlusive hydrocephalus exceeded a period of 14 days, that is, prior to the second exchange of EVDs. Weaning from the ventilator in patients who received this 413

3 H. B. Huttner, et al. therapy was generally started as soon as complications of ICH (hydrocephalus or aspiration pneumonia) were adequately treated. In cases of severe pneumonia, weaning was deferred until one of the following occurred: 1) FiO 2 levels were 40% or lower, to reach PO 2 levels of greater than 70 mm Hg; 2) the positive end-expiratory pressure was 8 cm H 2 O or lower; and 3) the ratio of inspiratory to expiratory time was 1:2 or lower. In all cases of protracted or unsuccessful weaning, tracheostomy was performed after 12 days of endotracheal intubation. Generally, patients were extubated as soon as GCS scores reached 8 and higher and in the presence of adequate brainstem reflexes. Pneumonia was diagnosed clinically, radiographically, and by increased infection parameters, and it was primarily treated using -lactam antibiotics or gyrase inhibitors. Patients were discharged from the ICU as soon as they were sufficiently weaned from the ventilator, cardiopulmonary function was stable, and they no longer required an EVD. Outcome Analysis A telephone interview was conducted with all surviving patients at the time of the study, or, if the patient had died, with the closest family members. Outcome at 1 year after onset of disease was determined. Patient function was evaluated using the mrs and the BI. 3,18,26 Excellent outcome was defined as a BI score of 90 to 100 and an mrs score of 0 to 1. Good outcome was defined as the ability to walk reasonably independently (mrs Score 2 3, BI Score 50 90). Poor outcome was defined as the inability to walk independently (mrs Score 4 5, BI Score 0 50) or having died (mrs Score 6). Statistical Analysis All statistical analyses were performed using a commercially available software package (Statview Version 5.0; SAS Institute, Inc., Cary, NC). A probability value of less than 0.05 was considered statistically significant. Normally distributed data are expressed as the means the SDs and were compared using the unpaired t-test. Nonnormally distributed data are expressed as median and range and were compared using the Mann Whitney U-test. The chi-square test was used for comparison of frequencies. Results Table 1 summarizes the characteristics in the patients in the study. Their mean age was years, and the mean hematoma volume was ml. To categorize the lesions as small or large hematomas, we subdivided patients into two groups; one contained patients with hematoma volumes larger than 25 ml, and one included patients with hematomas smaller than 25 ml. We also differentiated between a pure ganglionic hemorrhage and one with additional IVH and EVD (Table 2). Focusing on 40 patients with IVH and hydrocephalus, the median Graeb score was 7 (range 4 12), and EVDs were changed once in 24 patients (60%), twice in eight (20%), and a ventriculoperitoneal shunt was inserted in five (13%) (data not shown). The outcome analysis is given in Table 3 and Fig. 1. Patients with ganglionic hematomas smaller than 25 ml had significantly better outcomes than patients with hematoma TABLE 1 Hematoma volume, ICU stay, and other characteristics in 80 patients with ICH Characteristic volumes larger than 25 ml. No significant difference in the outcome could be found within these subgroups between patients with and without IVH and EVDs. A good recovery (mrs Score 1 3) was found in 62% of the patients with small hematomas, but in only 20% of the patients with large hematoma volumes. The overall mortality rate was 25%. A poor outcome (mrs Score 4 6) predominated in patients with large hematomas (80% compared with 38% in patients with small hematoma volumes; p 0.05, chi-square test). Of the patients with a poor outcome, 43% were dead after 1 year, but no significant difference could be found between patients with pure ganglionic hemorrhages and those with IVH and EVD. Finally, we analyzed the time spent in the ICU (Fig. 2). Patients with small hematomas required neurocritical care for a shorter period than patients with large ones (median 8 [range 2 19] days compared with 15 [range 5 29] days; p 0.05, Mann Whitney U-test). When comparing patients either with or without IVH and EVDs, the duration of the ICU stay was longer in cases of IVH, regardless of the hematoma volume, than in cases of pure ganglionic hematomas (median 8 [range 2 21] days compared with 16 [range 5 29] days; p 0.05, Mann Whitney U-test). Discussion Value* age (yrs) sex (F/M) 32:48 hematoma vol (ml) pure ganglionic ICH ganglionic ICH w/ IVH & EVD median Graeb score (range) 7 (4 12) days in ICU * All values except for the numbers of men and women in the study and the Graeb score are given as the means the SDs. There were 40 patients in each hematoma subgroup. In this case control study we compared the long-term outcome in patients with pure ganglionic hemorrhage with those who suffered from additional IVH and hydrocephalus. To avoid bias and to analyze a homogeneous patient contingent, we decided a priori to exclude those patients who had received no treatment. First, we found no significant differences in the long-term outcome between these groups when comparing matched pairs with similar hematoma volume and location, age, sex, GCS score on admission, and need for mechanical ventilation. Second, the outcome in general was more favorable in patients with smaller ( 25 ml) than in those with larger ( 25 ml) hematomas. Interestingly, this outcome was not influenced by the need for an EVD. Third, regarding the duration of stay in the ICU, patients who needed an EVD also required a longer period of neurocritical care in the ICU than those without IVH, regardless of the volume of the hematoma resulting from ICH. Many studies have been conducted to investigate the mortality rate and outcome in patients with ICH, IVH, and 414 J. Neurosurg. / Volume 105 / September, 2006

4 Influence of IVH and occlusive hydrocephalus on outcome of ICH TABLE 2 Characteristics and hematoma volumes of 80 patients in the matched pairs subgroups* Hematoma Vol (ml) Small ( 25) Large ( 25) Characteristic no IVH IVH & EVD no IVH IVH & EVD age (yrs) sex (F/M) 6:14 9:11 7:13 10:10 hematoma vol median Graeb score NA 7 NA 7 thalamic location (%) * There were 40 patients in each size category, in subgroups of 20 each classified as pure ganglionic hematomas and lesions with IVHs that required EVDs. The cut-off hematoma volume was chosen because of the mean lesion volume in all patients, which was approximately 25 ml. Note that the distribution of thalamic hematomas was similar for small and large lesions, and that the median Graeb score was identical. Values for patient age and hematoma volume are given as the means SDs. Abbreviation: NA = not applicable. TABLE 3 Outcome of patients with large compared with small ganglionic hematomas* Hematoma Size (ml) Small ( 25) Large ( 25) Outcome Parameter no IVH IVH & EVD no IVH IVH & EVD excellent good poor dead * Categorized according to the presence of IVH necessitating an EVD. Excellent outcome, BI /mRS Score 0 1; good outcome, BI 50 90/mRS Score 2 3; poor outcome, BI 0 50/mRS Score 4 5; dead, mrs Score 6. The incidence of mortality was higher in patients with large compared with small hematomas (p 0.05, Mann Whitney U-test). Within the groups of small and large hematomas, no significant differences were found for patients with or without IVH requiring EVD. hydrocephalus. Generally, the prognosis for patients with IVH is affected by several factors. A rise in ICP associated with initial bleeding may cause a significant reduction in cerebral blood flow that potentially leads to additional ischemia. 19,20 Obstruction of CSF flow and the mass effect associated with clots in the satellite hematoma and within the ventricles may produce further damage. In previous studies it has been demonstrated that the presence of IVH as well as the presence of obstructive hydrocephalus are independent predictors of a poor outcome. 21,22,28 When only parts of the ventricular system are affected, the reported patient mortality rate (32 44%) does not vary significantly from that of patients with purely ganglionic hemorrhages. 6,12,28 Nevertheless, in cases of hemorrhage extension into all four ventricles with subsequent hydrocephalus, the mortality rate increases up to 60 to 90%, depending on the hematoma volume and location. 8,12,22,28 Moreover, morbidity and mortality rates seem to be influenced by the extent of the intraventricular bleeding and by the location of the intracerebral hematoma; a thalamic location has the worst prognosis. 58,28,31 In our study, however, the frequency of thalamic bleeding was similar in the comparison groups of small and large hematomas, and due to the matching process, and to avoid bias identical within them. In these previous studies investigators mainly analyzed outcome after 3 months. Interestingly, analysis of our data shows a similar long-term outcome after 12 months when comparing purely ganglionic hematomas with those involv- FIG. 1. Bar graph showing long-term outcome in patients with ganglionic hematomas. The difference between patients with small and large hematomas was significant (p 0.05, Mann Whitney U-test). Note that within the subgroups of patients with small and large hematomas, no differences in the long-term outcome could be found between patients with pure ganglionic hemorrhages and those with additional IVH and EVDs. J. Neurosurg. / Volume 105 / September,

5 H. B. Huttner, et al. FIG. 2. Graph showing the duration of stay in the ICU of all patients, grouped by small and large hematomas and by the presence of IVH. Data are presented as the median (diamonds) and range. Note that the duration of stay in the ICU was longer in patients who required an EVD than in patients with purely ganglionic hemorrhages, regardless of hematoma volume. ing additional IVH that required an EVD. Possible explanations include the severity of the disease that is causing the ICH, which in itself may lead to an unfavorable long-term outcome. Our data revealed no patient with an excellent outcome. The overall poor outcome does not seem to be worsened by additional hemorrhage, even when it affects the fourth ventricle and results in occlusive hydrocephalus. 13 Indeed, the hematoma volume seems to be more important in influencing the long-term outcome than is the presence of IVH or hydrocephalus. However, it must be mentioned that not every patient with IVH and hydrocephalus was analyzed; we focused on the patients who received treatment. An interesting observation concerned the duration of ICU treatment of the patients included in this study. Although it was not surprising that patients with larger hematoma volumes from ICH required critical care medicine for a longer period than did patients with smaller lesions, we observed an extensively prolonged ICU stay in patients with hydrocephalus due to IVH who required an EVD. In any case, our patients were matched even with regard to the need for mechanical ventilation. The temporary sedation and intubation necessary for surgical insertion of the EVD, with the adverse effects of intubation, such as protracted weaning due to concurrent cardiopulmonary diseases or respiratorassociated pneumonia, 14 cannot be posited as the possible cause of a prolonged ICU stay. Rather, the presence of IVH itself, with the consequent need for ventriculostomy, remains the primary reason for requiring prolonged neurocritical care therapy. Although they have a relatively low prevalence, infectious agents in the EVD are known to occur despite administration of antibiotic medications in the course of treatment. 1,17 Furthermore, CSF infections in the presence of ventricular catheters lengthen the ICU stay, whereas they have not been shown to have a significant impact on the clinical outcome. 30 Our study has limitations because the results were obtained retrospectively by analysis of a selected patient contingent, and because they reflect single-center treatment regimens. The three major limitations of our study are here described in more detail. First, because the patients with pure ganglionic hematomas who were included in the study were selected by matching them with patients with IVH plus hydrocephalus and EVD, it should be noted that, according to institutional protocols for the treatment of severe ganglionic hemorrhages, only patients with a reasonably favorable prognosis underwent placement of an EVD. Our mean mortality rate was only 25%, and therefore this result cannot be conclusively compared with the studies cited, because our patients presumably reflect a putative group of individuals with ICH whose initial evaluation yields a favorable prognosis. In addition, we did not enroll patients who received intraventricular fibrinolysis. Taking these excluded patients into account, the results of this study might not be applicable to all patients with ICH. Second, we chose to perform a clinical evaluation after 12 months to determine the long-term outcome. This time point is uncommonly long after onset of the disease and might be affected by fac- 416 J. Neurosurg. / Volume 105 / September, 2006

6 Influence of IVH and occlusive hydrocephalus on outcome of ICH tors such as comorbidity that are not related to ICH. Third, with regard to the finding that patients with EVDs had a prolonged stay in the ICU, other variables besides the requirement of an EVD might affect the duration of neurocritical care needed. These possible influencing factors have not been investigated sufficiently in this study. Definitive statements about the prognosis and treatment regimens in patients with ganglionic hemorrhage, a disease that might be complicated by IVH and hydrocephalus, will not be available until prospectively randomized trials focus on this concern. Conclusions When focusing on treated patients, the overall long-term outcome of individuals with additional IVH and hydrocephalus compared with patients who had purely ganglionic hemorrhage was similar, despite the longer stay in the ICU. Involvement of the ventricular system and the presence of hydrocephalus are known risk factors for increased mortality rates and poor outcome. 7,19,24,28 These complications, however, when being treated appropriately in individuals with a reasonably favorable prognosis, do not seem to influence the long-term outcome in patients with deep ganglionic ICH. References 1. Adams RE, Diringer MN: Response to external ventricular drainage in spontaneous intracerebral hemorrhage with hydrocephalus. Neurology 50: , Andrews CO, Engelhard HH: Fibrinolytic therapy in intraventricular hemorrhage. Ann Pharmacother 35: , Bonita R, Beaglehole R: Recovery of motor function after stroke. Stroke 19: , Cheung RTF, Zou LY: Use of the original, modified, or new intracerebral hemorrhage score to predict mortality and morbidity after intracerebral hemorrhage. Stroke 34: , Darby DG, Donnan GA, Saling MA, Walsh KW, Bladin PF: Primary intraventricular hemorrhage: clinical and neuropsychological findings in a prospective stroke series. Neurology 38:68 75, Daverat P, Castel JP, Dartigues JF, Orgogozo JM: Death and functional outcome after spontaneous intracerebral hemorrhage. A prospective study of 166 cases using multivariate analysis. Stroke 22:1 6, Diringer MN, Edwards DF, Zazulia AR: Hydrocephalus: a previously unrecognized predictor of poor outcome from supratentorial intracerebral hemorrhage. Stroke 29: , Engelhard HH, Andrews CO, Slavin KV, Charbel FT: Current management of intraventricular hemorrhage. Surg Neurol 60: 15 21, Gebel JM, Broderick JP: Intracerebral hemorrhage. Neurol Clin 18: , Gebel JM, Sila CA, Sloan MA, Granger CB, Weisenberger JP, Green CL, et al: Comparison of the ABC/2 estimation technique to computer-assisted volumetric analysis of intraparenchymal and subdural hematomas complicating the GUSTO-1 trial. Stroke 29: , Graeb DA, Robertson WD, Lapointe JS, Nugent RA, Harrison PB: Computed tomographic diagnosis of intraventricular hemorrhage. Etiology and prognosis. Radiology 143:91 96, Jichici D, Frank JI: Thrombolytic therapy in neurointensive care. Crit Care Clin 13: , Juvela S: Risk factors for impaired outcome after spontaneous intracerebral hemorrhage. Arch Neurol 52: , 1995 J. Neurosurg. / Volume 105 / September, Klopfenstein JD, Kim LJ, Feiz-Erfan I, Hott JS, Goslar P, Zabramski JM, et al: Comparison of rapid and gradual weaning from external ventricular drainage in patients with aneurysmal subarachnoid hemorrhage: a prospective randomized trial. J Neurosurg 100: , Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al: The ABCs of measuring intracerebral hemorrhage volumes. Stroke 27: , Liliang PC, Liang CL, Lu CH, Chang HW, Cheng CH, Lee TC, et al: Hypertensive caudate hemorrhage prognostic predictor, outcome, and role of external ventricular drainage. Stroke 32: , Lozier AP, Sciacca RR, Romagnoli MF, Connolly ES Jr: Ventriculostomy-related infections: a critical review of the literature. Neurosurgery 51: , Mahoney FI, Barthel DW: Functional evaluation: the Barthel index. Md State Med J 14:61 65, Marti-Fabregas J, Piles S, Guardia E, Marti-Vilalta JL: Spontaneous primary intraventricular hemorrhage: clinical data, etiology and outcome. J Neurol 246: , Mayer SA, Lignelli A, Fink ME, Kessler DB, Thomas CE, Swarup R, et al: Perilesional blood flow and edema formation in acute intracerebral hemorrhage: a SPECT study. Stroke 29: , Mayfrank L, Kissler J, Raoofi R, Delsing P, Weis J, Kuker W, et al: Ventricular dilatation in experimental intraventricular hemorrhage in pigs: characterization of cerebrospinal fluid dynamics and the effects of fibrinolytic treatment. Stroke 28: , Mayfrank L, Lippitz B, Groth M, Bertalanffy H, Gilsbach JM: Effect of recombinant tissue plasminogen activator on clot lysis and ventricular dilatation in the treatment of severe intraventricular haemorrhage. Acta Neurochir 122:32 38, Nilsson OG, Lindgren A, Brandt L, Säveland H: Prediction of death in patients with primary intracerebral hemorrhage: a prospective study of a defined population. J Neurosurg 97: , Phan TG, Koh M, Vierkant RA, Wijdicks EFM: Hydrocephalus is a determinant of early mortality in putaminal hemorrhage. Stroke 31: , Qureshi AI, Tuhrim S, Broderick JP, Batjer HH, Hondo H, Hanley DF: Spontaneous intracerebral hemorrhage. N Engl J Med 344: , Rankin J: Cerebral vascular accidents in patients over the age of 60. I. General considerations. Scott Med J 2: , Tuhrim S, Dambrosia JM, Price TR, Mohr JP, Wolf PA, Heyman A, et al: Prediction of intracerebral hemorrhage survival. Ann Neurol 24: , Tuhrim S, Horowitz DR, Sacher M, Godbold JH: Volume of ventricular blood is an important determinant of outcome in supratentorial intracerebral hemorrhage. Crit Care Med 27: , van Gijn J, Hijdra A, Wijdicks EFM, Vermeulen M, van Crevel H: Acute hydrocephalus after aneurysmal subarachnoid hemorrhage. J Neurosurg 63: , Wong GK, Poon WS, Wai S, Yu LM, Lyon D, Lam JM: Failure of regular external ventricular drain exchange to reduce cerebrospinal fluid infection: result of a randomised controlled trial. J Neurol Neurosurg Psychiatry 73: , Zurasky JA, Aiyagari V, Zazulia AR, Shackelford A, Diringer MN: Early mortality following spontaneous intracerebral hemorrhage. Neurology 64: , 2005 Manuscript received July 24, Accepted in final form March 2, Address reprint requests to: Hagen B. Huttner, M.D., Department of Neurology, University of Heidelberg, INF 400, Heidelberg, Germany. hagen.huttner@neuro.imed.uni-erlangen.de. 417