Introduction 

   Deep intracerebral hemorrhage (ICH) is critical disease because it has higher morbidity and mortality. There is controversy in management of deep ICH, medical conservative care and surgical intervention.⁷⁾ Surgical evacuation of hematoma is effective in reduction of hematoma volume, lowering intracranial pressure, and recovery of perfusion in cerebral hemisphere.^2)4)11) But open craniotomy causes additional damage to overlying brain parenchyma. Catheter insertion into the deep ICH is less invasive and cause little damage to brain parenchyma. This method has been assisted by stereotaxy, such as frame based stereotactic technique and its accuracy has been improved.¹⁰⁾ In recent years, frameless stereotactic method has been introduced and used world widely. However its accuracy has been questioned. The aim of this study is to evaluate accuracy of frameless stereotactic technique for deep ICH catheterization and its result. 

Patients and Methods 

Patients 
   From January 2006 to November 2006, we identified 27 patients who were diagnosed as deep ICH and underwent navigation-assisted frameless stereotactic catheter insertion with/without thrombolysis by urokinase irrigation. Clinical and radiologic data included the following: age, sex, initial Glasgow coma scale (GCS), preoperative hematoma location and volume, coexistence of intraventricular hemorrhage (IVH). 

Hematoma volume measurement 
   Hematoma volume was calculated according to the ABC method had been described by Kothari et al.⁶⁾ Digitalized CT image was measured on GE PACS system. 

Techniques of navigation-assisted frameless stereotatic catheter insertion 
   Fiducial markers were patched around the entry point, usually Kocher's point or above the subcortical hematoma. 1 mm slice non-contrast CT scan was taken for navigation system. The images were reconstructed in the system (CART-II, Stryker Instuments, Kalamazoo, MI) and matched its points to fiducial markers on scalp of the patients. Target point was created at the center of hematoma. Catheter insertion was done under general anesthesia, with head fixation by the Mayfield head rest. After aseptic preparation, scalp was incised straightly. Small hole was drilled and dura was opened. The hole was usually located at Kocher's point or the closest one to the target point. A catheter was inserted toward target using the guiding instruments. Hematoma was aspirated with low tension by syringe about 5 to 10 ml due to target confirmation and acute decompression of mass effect. 

Post operative care and outcome assessment 
   All patients underwent CT study immediately after operation. It was compared with preoperative image, and catheter tip location was calculated by Euclidean x-y-z method. 
   Subsequent thrombolysis and clot aspiration were performed at the bedside using sterile technique. Urokinase 3000 IU in 1 ml of preservative-free saline was injected into the catheter. After 6 to 8 hours, manual aspiration of lysed clot was attempted at the bedside, and the aspirated volume was recorded. Catheter was preserved until drainage below 10cc/day or CSF drained through a catheter due to communication between ventricle and hematoma. In patients whose catheter was located deeper than hematoma center, it has been withdrawn. 7 days later from catheter removal, CT scan was followed. 
   Final outcome of patient was scaled by Glasgow outcome scale (GOS) at 6 month after the operation. 

Statistical analysis 
   Independent T-test, ANOVA test, and univariate analysis were used for comparison between target accuracy and hematoma location and size. Paired T-test was used for comparison between preoperative hematoma volume and final hematoma volume. Multivariate regression analysis was performed to analyze the prognostic impact of the following parameters on GOS: gender, age, preoperative clinical status, hematoma volume before and after treatment, localization, ventricular extension. For that purpose, the GOS was dichotomized into favorable (GOS>3) and unfavorable outcome (GOS≤3). The level of significance was set at 0.05. 

Results 

   Clinical characteristics and treatment of patients with deep ICH treated by navigation-assisted aspiration and thrombolysis was summarized in Table 1. 
   There were 20 men and 7 women, and the age ranged between 40 and 90 years (mean, 58.3±13.2 years). The initial GCS score ranged between 7 and 14 with a mean of 11.5. Hematoma volume ranged 20 ml and 50 ml (mean 34 ml). ICH was located in basal ganglia with 8 patients, thalamus with 10 patients, and subcortical with 9 patients. Nine patients had IVH simultaneously. Rebleeding before operation occurred in one patient (patient 18). Navigation CT showed the increased ICH volume when compared one in the initial CT scan. 
   Technical purpose of surgery was that catheter tip was located to the center of ICH. The length between center of hematoma and tip of catheter was calculated. Its mean length was 6.8 mm (range between 0 and 15 mm). Catheter tip and target matched in 8 patients (29.6%). In cases of inappropriately located catheter tip (70.4%), most of hematomas were thalamic in location due to long trajectory (9 of 10 thalamic locations), which were statistically significant (p<0.05). However, hematoma size was not significantly correlated with accuracy. Also, univariate analysis showed no statistical significance between these two factors and accuracy. 
   There were no postoperative complications in all patients. Postoperative urokinase irrigation was done in 16 patients. Bleeding that occurred after urokinase irrigation was not developed in any cases. 
   Remnant ICH was measured in final CT scan before removal of catheter. They were taken almost 3rd post operative day. The remnant ICH was detected in 22 patients. Initial ICH volume was reduced by an average of 83% (range 40 to 100%) and average final hematoma volume was 5.9 ml (range 0 to 18 ml). Preoperative hematoma volume had statistically significant correlation with final hematoma volume (p<0.05). Mean GOS points was 3.4. No mortality was reported. In multiple regression analysis, initial GCS score was statistically significant (Table 2, p<0.05). However, the preoperative hematoma volume, final hematoma volume, and hematoma location were not significant. 

Discussion 

   This report with the preliminary results showed that navigation-assisted frameless stereotactic ICH catheterization has limited accuracy but effectiveness for management choice of ICH volume reduction. 
   ICH catheterization and thromobolysis had been studied by several investigators.^{9)10)14)15)17)} However they were failed to prove ICH aspiration and thrombolysis is superior to medical treatment, so it still remains controversial. Catheter insertion is suggested as one of several treatment modality of ICH. In recent years, several studies were processed with the frameless stereotactic navigation, because of its advantages superior to the frame based.¹³⁾¹⁶⁾ 
   Frameless navigation operation does not need to frame fixation before anesthesia, so patient does not suffer from head fixation. Rebleeding was less frequently occurred in the frameless. Thiex et al. reported that rebleeding rate of frameless stereotaxy was 10.7%, frame based stereotaxy was 17%.¹⁶⁾ Only one case underwent rebleeding event in our study, it was occurred in preoperative period. Stereotactic frame appliance is more complicated and painful than fiducial markers setting on scalp. It can increase blood pressure, and finally cause rebleeding. 
   Accuracy of the frameless stereotaxy had been proved also in brain biopsy.¹²⁾ However, frame based stereotaxy is more accurate than frameless. In frameless stereotaxy, fiducial markers are patched on the scalp, so it is movable by stretching of scalp. Even though some study reported that head position during preoperative imaging did not affect on error of navigation procedure,¹²⁾ markers on the scalp could be displaced during CT scanning or fixation of skull, and finally it serves increasing error. Dorward et al. reported that error of CT image based frameless stereotactic biopsy is 4.8 mm.³⁾ In our study, mean error was 6.8 mm. This difference may be due to usage of locking instrument holder. For more accurate result, holding device should be needed during catheter advancement toward target.⁵⁾ Malposition can be detected during operation by blood clot drainage. If fiducial markers preserved during operation, postoperative image could be used for re-operation data. It is also possible in frame based surgery. Thiex et al. reported other complication of catherization by both methods, there was no significant difference in infection rate.¹⁶⁾ 
   Frame-based stereotaxy is time consuming.¹⁾ Even if we did not assess the time it takes to utilize frameless stereotaxy in the management of ICH, we assume that this technique is less time-consuming. After performing CT scan, we could transfer the patient directly to the operating room for frameless stereotactic catheter placement. The required time for frameless stereotaxy is further reduced by direct 3- dimensional visualization of possible trajectories without previous target point selection and data processing. If comparing frame-based and frameless stereotaxy for deep ICH management, the duration of each procedure should be taken into account, because at least in animal models, the degree of secondary tissue damage in the vicinity of the ICH was related to the quickness of hematoma evacuation.¹⁶⁾ 
   There have been a small series of frameless stereotaxy in the management of ICH,¹³⁾¹⁶⁾ but data of frame-based stereotaxy for hematoma puncture and subsequent thrombolysis are available.¹⁰⁾¹⁵⁾ Increasing age, and greater severity either measured clinically (GCS, limb paresis) or by imaging (hematoma volume, intraventricular extension) are usually indicated as poor prognostic factors.⁸⁾ Residual volume of ICH after therapy or percent change of ICH volume from baseline is also important for functional outcome.¹⁰⁾ In our study, the only significant factor affecting the prognosis was the preoperative neurologic status. The present study is limited due to small number of patients and various hematoma location. Further investigations are required of larger number of patients and controlled design for validating frameless stereotactic catheter insertion and subsequent thrombolysis. 

Conclusion 

   Navigation-assisted frameless stereotactic ICH catheterization has limited accuracy but effectiveness for management choice of ICH volume reduction. There were no postoperative complications in all patients. The only significant factor affecting the prognosis was the preoperative neurologic status. 

REFERENCES

1. Barnett GH, Kormos DW, Steiner CP, Weisenberger J. Intraoperative localization using an armless, frameless stereotactic wand. J Neurosurg 78:510-4, 1993 

2. Belayev L, Saul I, Curbelo K, Busto R, Belayev A, Zhang Y, Riyamongkol P, Zhao W, Ginsberg MD. Experimental intracerebral hemorrhage in the mouse: histological, behavioral, and hemodynamic characterization of a doubleinjection model. Stroke 34:2221-7, 2003 

3. Dorward NL, Alberti O, Palmer JD, Kitchen ND, Thomas DG. Accuracy of true frameless stereotaxy: in vivo measurement and laboratory phantom studies. Technical note. J Neurosurg 90:160-8, 1999 

4. Grotta JC. Management of Primary Hypertensive Hemorrhage of the Brain. Curr Treat Options Neurol 6:435-42, 2004 

5. Grunert P, Espinosa J, Busert C, Gunthner M, Filippi R, Farag S, Hopf N. Stereotactic biopsies guided by an optical navigation system: technique and clinical experience. Minim Invasive Neurosurg 45:11-5, 2002 

6. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, Khoury J. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 27:1304-5, 1996 

7. Little KM, Alexander MJ. Medical versus surgical therapy for spontaneous intracranial hemorrhage. Neurosurg Clin N Am 13:339-47, 2002 

8. Lisk DR, Pasteur W, Rhoades H, Putnam RD, Grotta JC. Early presentation of hemispheric intracerebral hemorrhage: prediction of outcome and guidelines for treatment allocation. Neurology 44:133-9, 1994 

9. Miller CM, Vespa PM, McArthur DL, Hirt D, Etchepare M. Frameless stereotactic aspiration and thrombolysis of deep intracerebral hemorrhage is associated with reduced levels of extracellular cerebral glutamate and unchanged lactate pyruvate ratios. Neurocrit Care 6:22-9, 2007 

10. Montes JM, Wong JH, Fayad PB, Awad IA. Stereotactic computed tomographic-guided aspiration and thrombolysis of intracerebral hematoma: protocol and preliminary experience. Stroke 31:834-40, 2000 

11. Proust F, Leveque S, Derrey S, Tollard E, Vandhuick O, Clavier E, Langlois O, Freger P. Spontaneous supratentorial cerebral hemorrhage: role of surgical treatment. Neurochirurgie 53:58-65, 2007 

12. Reinges MH, Krings T, Nguyen HH, Hans FJ, Korinth MC, Holler M, Kuker W, Thiex R, Spetzger U, Gilsbach JM. Is the head position during preoperative image data acquisition essential for the accuracy of navigated brain tumor surgery? Comput Aided Surg 5:426-32, 2000 

13. Rohde V, Rohde I, Reinges MHT, Mayfrank L, Gilsbach JM. Frameless stereotactically guided catheter placement and fibrinolytic therapy for spontaneous intracerebral hematomas: Technical aspects and initial clinical results. Minim Invas Neurosurg 43:9-17, 2000 

14. Schaller C, Rohde V, Hassler W. Local thrombolytic treatment of spontaneous intracerebral hemorrhage with plasminogen activator (rt-PA). Indications and limits. Nervenarzt 66:275-81, 1995 

15. Teernstra OP, Evers SM, Lodder J, Leffers P, Franke CL, Blaauw G. Stereotactic treatment of intracerebral hematoma by means of a plasminogen activator: a multicenter randomized controlled trial (SICHPA). Stroke 34:968-74, 2003 

16. Thiex R, Rohde V, Rohde I, Mayfrank L, Zeki Z, Thron A, Gilsbach JM, Uhl E. Frame-based and frameless stereotactic hematoma puncture and subsequent fibrinolytic therapy for the treatment of spontaneous intracerebral hemorrhage. J Neurol 251:1443-50, 2004 

17. Vespa P, McArthur D, Miller C, O'Phelan K, Frazee J, Kidwell C, Saver J, Starkman S, Martin N. Frameless stereotactic aspiration and thrombolysis of deep intracerebral hemorrhage is associated with reduction of hemorrhage volume and neurological improvement. Neurocrit Care 2:274-81, 2005