Diagnostic Accuracy and Field for Improvement of Frameless Stereotactic Brain Biopsy: A Focus on Nondiagnostic Cases

Abstract

Background The diagnostic accuracy of frameless stereotactic brain biopsy has been reported, but there is limited literature focusing on the reasons for nondiagnostic cases. In this study, we evaluate the diagnostic accuracy of frameless stereotactic brain biopsy, compare it with the current international standard, and review the field for improvement.

Methods This is a retrospective analysis of consecutive, prospectively collected frameless stereotactic brain biopsies from 2007 to 2020. We evaluated the diagnostic accuracy of the frameless stereotactic brain biopsies using defined criteria. The biopsy result was classified as conclusive, inconclusive, or negative, based on the pathologic, radiologic, and clinical diagnosis concordance. For inconclusive or negative results, we further evaluated the preoperative planning and postoperative imaging to review the errors. A literature review for the diagnostic accuracy of frameless stereotactic biopsy was performed for the validity of our results.

Results There were 106 patients with 109 biopsies performed from 2007 to 2020. The conclusive diagnosis was reached in 103 (94.5%) procedures. An inconclusive diagnosis was noted in four (3.7%) procedures and the biopsy was negative in two (1.9%) procedures. Symptomatic hemorrhage occurred in one patient (0.9%). There was no mortality in our series. Registration error (RE) and inaccurate targeting occurred in three trigonal lesions (2.8%), sampling of the nonrepresentative part of the lesion occurred in two cases (1.8%), and one biopsy (0.9%) for lymphoma was negative due to steroid treatment. The literature review suggested that our diagnostic accuracy was comparable with the published literature.

Conclusion The frameless stereotactic biopsy is a safe procedure with high diagnostic accuracy only if meticulous preoperative planning and careful intraoperative registration is performed. The common pitfalls precluding a conclusive diagnosis are RE and biopsies at nonrepresentative sites.

Keywords

frameless stereotaxy - histology - image-guided surgery - intracranial - neuronavigation

Background

Stereotactic brain biopsy is widely used for diagnosing intracranial lesions not amenable to surgical excision. Both frame-based and frameless techniques have been validated for their efficacies. Previous research has suggested that there were no significant differences in diagnostic yield and complication rate between frame-based and frameless procedures.[1] [2] However, there is considerable heterogeneity in the definition of diagnostic accuracy or diagnostic yield, and limited literature is published on the actual causes of the nondiagnostic cases. Frameless stereotactic biopsy has been performed in our unit since 2007. A lower hemorrhagic rate using a 1.8-mm-diameter biopsy needle compared with a 2.5-mm-diameter biopsy needle was previously reported in our series of 54 procedures.[3] We have been adhering to the same protocol of frameless stereotactic biopsy since then.[4] In this article, we reviewed consecutive frameless stereotactic procedures performed in a 14-year interval with an aim to analyze the errors being responsible for nondiagnostic cases and to scrutinize the field for improvements.

Methods

This is a single-center case series for consecutive patients who underwent frameless stereotactic brain biopsy in a tertiary neurosurgical center in Hong Kong. All patients who underwent frameless stereotactic brain biopsy from 2007 to 2020 in our institute were included. Patients with frame-based stereotactic biopsy or patients with intraoperative ultrasound-guided biopsy were excluded. Patient demographics, including age at operation, gender, and medical comorbidities, were recorded. The institutional ethics committee approved the study protocol with patient consent waived as this study was a secondary analysis of available data with no additional risk to patients. The report of this case series adhered to the preferred reporting of case series in surgery (process) guideline.[5]

Surgical Indication and Preoperative Preparations

The clinical history, physical signs, investigations, and imaging findings of all biopsy cases were discussed in the preoperative meeting with neurosurgeons, neuroradiologists, and neurologists. The indications for brain biopsies were lesions in which the pathologic diagnosis would be essential for the subsequent treatment. For a neoplastic lesion, a biopsy was indicated if the lesion cannot be treated by surgical excision, that is, gross total resection without neurologic deficit was impossible and survival was not improved by partial resection.[6] This type of lesion was commonly deep seated, infiltrative without a margin, or in the eloquent cortex. For a lesion with imaging findings not suggestive of a neoplastic lesion, a biopsy was indicated for representative sampling for further microbiological investigation or pathologic delineation.[6] Antiplatelets and anticoagulation therapy were discontinued before the operation.

Target Planning

The target planning had been described previously by Yuen et al.[3] Preoperative images including computed tomography (CT) and magnetic resonance imaging (MRI) of the brain were fused. In general, the target of the biopsy was set at the center of the most representative part of the lesion, considering the contrast enhancement, location of the lesion, and possible vasculature. A spectroscopy study was also used for reference if available. If there were multiple lesions, the most representative lesion with the lowest possibility of causing a biopsy-associated neurologic deficit was chosen for biopsy. The actual biopsy would be taken at the target site first, then in different directions at the target site, and subsequently deeper or more superficial than the target along the trajectory if necessary. The entry site and trajectory were planned considering the following factors: (1) gyral entry; (2) shortest distance to the target; (3) avoidance of sulci, vessels, ventricles, and critical functional areas such as the motor cortex and the internal capsule; and (4) aligning the biopsy window along the long axis of the lesion, especially for small oval shape lesions. The target lesion size was defined as the dimension of the lesion measured on the probe view at the target level and it was grouped under <10 or ≥10 mm. The target locations were classified into (1) cerebrum; (2) insula, basal ganglia, and thalamus; (3) pineal and brainstem; and (4) cerebellum. The insular, basal ganglia, thalamus, pineal, and brainstem lesion were considered deep lesions. Each biopsy trajectory was carefully reviewed with consideration of the accuracy of the VarioGuide system (BrainLAB AG, Feldkirchen, Germany) as previously reported by the phantom-based measurements. The scalar error (mean ± SD) of the biopsy tip to the defined target point was 1.44 ± 0.98 mm in a sample of 129 measurements.[7] The upper margin of the 99.9% confidence interval of the scalar error was 1.72 mm. To achieve a safe biopsy, the outer rim of a 1.8-mm biopsy needle should be located inside the target as well, which requires another 0.9-mm margin on top of a 1.72-mm error (i.e., 2.62-mm margin from a target). Therefore, we considered that under accurate registration, the frameless biopsy was feasible only when no critical structure was within the 2.62-mm radius of the biopsy trajectory. This requirement may not be achieved for a proportion of patients with a small lesion of less than 10 mm in the pineal region or brainstem. Frame-based stereotactic biopsy was considered should the frameless biopsy be not feasible.

Operative Procedure

All the included biopsies were performed under general anesthesia using frameless stereotaxy. BrainLAB navigation (BrainLAB AG, Feldkirchen, Germany) together with the instrument holder, VarioGuide system (BrainLAB AG, Feldkirchen, Germany) were used for the majority of the cases, and Fiagon navigation system (Fiagon GmbH, Hennigsdorf, Germany) were used when BrainLAB navigation was not available. Prophylactic antibiotic was routinely given. The majority of patients were positioned in the supine or supine-oblique (with the ipsilateral shoulder raised by a pad plus head rotation to the contralateral side) position, while few cases were performed in the lateral position.

The registration method was surfacing matching. A spiral CT of 0.625-mm slice interval or a T1-3D MRI less than 1 mm per slice was used as the image source for registration. The accuracy of registration was checked meticulously using the software's function and at the anatomical landmarks including nasion, lateral and medial orbital canthi, tragus, and scalp surface.

After a burr hole was made, the dura was opened to confirm gyral entry. The biopsy needle was then inserted through the instrument holder toward the target under navigation guidance. Upon reaching the planned target, a syringe was applied to the inner cannula of the biopsy needle, and the side window was opened. Negative pressure was applied through a 2.5-mL syringe routinely unless it was insufficient to obtain the specimen. In contrast, a 1-mL syringe was used for targets located in the brainstem. To obtain an adequate amount of tissue and to increase the sampling area, specimens were taken in different directions by rotating the needle plus minor moving of the needle back and forth along the same trajectory.

In all the cases, the specimen was sent for the intraoperative frozen section or smear. Direct communication was made intraoperatively with the neuropathologist to discuss the histologic result and specimen adequacy for subsequent pathologic study. Additional specimens would be obtained from different depths and orientations if the intraoperative histologic result was uncertain. An additional amount of specimen would also be obtained if the remaining specimen was anticipated insufficient for subsequent pathologic study, provided that no intraoperative hemorrhage was encountered, that is, blood was aspirated or came out from the lumen of the biopsy needle. From 2007 to 2010, a 2.5-mm-diameter Sedan-type side-cutting biopsy needle (Elekta AB, Stockholm, Sweden) was used, and after 2010, we changed to a 1.8-mm-diameter biopsy needle (BrainLAB AG, Feldkirchen, Germany).[3]

Postoperative Imaging

A spiral CT of 0.625-mm slice interval was performed within 12 hours after surgery in each case and was fused to the preoperative imaging for correlation of the biopsy site. The planned target was evaluated for being inside the actual biopsy site (within the air bubbles on postoperative CT or not), any visible biopsy tract, or hemorrhage at the biopsy site or along the tract.

Diagnostic Accuracy

The final diagnosis was discussed in a regularly held combined meeting with neurosurgeons, neuropathologists, radiologists, and neurologists. The biopsy result was classified as (1) conclusive, (2) inconclusive, or (3) negative. The criteria for each diagnostic category are shown in [Table 1]. Diagnostic accuracy represented the percentage of cases with a conclusive diagnosis. A conclusive diagnosis was defined as (1) tumor, that is, confirmed tumor type and the WHO grading; (2) infection, that is, confirmed infection and microorganism identified; (3) neurologic disorders, that is, characteristic abnormal tissue was obtained, the target site was accurate on postoperative imaging, and the histologic findings were consistent with the clinical course of the neurologic disorders and other systemic investigations. An inconclusive diagnosis was defined as the following: abnormal tissue was obtained but only a nonspecific descriptive diagnosis could be made. If no characteristic abnormal tissue was obtained but only a descriptive histologic diagnosis was made (i.e., necrosis), and the subsequent open biopsy confirmed the same result, the stereotactic biopsy would remain inconclusive. A negative biopsy was defined as the following: only normal brain tissue was obtained and/or the histologic diagnosis was contradictory to a subsequent biopsy, other investigations, or the clinical course of the disease. Cases with inconclusive diagnoses and negative biopsies were considered nondiagnostic. The completeness of an immunohistochemical (IHC) study and the agreement between biopsy and open surgery results were also evaluated. The preoperative planning, the operation, and the postoperative imaging analysis were supervised by the same senior consultant for all procedures.

Table 1 Criteria for the diagnostic accuracy by category in the current study

Category	Tissue	Histologic	Radiologic	Clinical
Conclusive	Pathologic	Tumor[a]	Accurate	Consistent
		Infection[b]	Accurate	Consistent
		Neurologic disorder[c]	Accurate	Consistent
Inconclusive	Abnormal	Nonspecific descriptive diagnosis	Accurate or deviated	Consistent
Negative	Nondiagnostic	Normal brain, blood clots, etc.	Not specified	Contradicting

^a Tumor: histologic type and grading confirmed.

^b Infection: microorganism identified or confirmed by characteristic histologic findings.

^c Neurologic disorder: characteristic abnormal tissue obtained.

Diagnosis of Postoperative Hemorrhage

Any visible new hyperdense signals were strictly documented as hemorrhages, with the largest diameter measured on the CT images.[3] The size of the hemorrhage was grouped into less than 5, 5 to 10, and greater than 10 mm. The hemorrhage was classified into symptomatic and asymptomatic according to any associated neurologic deterioration.

Literature Review

A literature review was performed using PubMed and Medline (OvidSP) for the diagnostic accuracy or the diagnostic yield of frameless stereotactic brain biopsy for patients with intracranial lesions. We aimed to evaluate the original studies with case numbers of more than 100. The detailed syntaxes are shown in [Table 2].

Table 2 The searching method and syntax for the literature review on the diagnostic accuracy of frameless stereotactic biopsy

Date of query	June 4, 2022	Database
Field	Syntax	PubMed	Ovid MEDLINE
Domain	#1 Brain or intracranial or cerebral	All fields	Keyword (map term to subject heading)
Determinant	#2 Frameless stereotactic brain biops[*] or frameless stereotactic biops[*] or image-guided brain biops[*] or image-guided biops[*] or stereotactic brain biops[*] or stereotactic biops[*]	Title/abstract	Keyword (map term to subject heading)
Outcome	#3 Diagnostic yield or diagnostic rate or diagnostic accuracy or accuracy or nondiagnostic yield or diagnostic error or nondiagnostic biops[*]	Title/abstract	Keyword (map term to subject heading)
Search		#1 and #2 and #3	#1 and #2 and #3
Filter		From 2011 to 2022	From 2011 to 2022

^* In this table the "*" symbol is part of the searching syntaxes used to present searching Biopsy/Biopsies in at the same time.

Statistical Analysis

The statistical analyses were performed by IBM SPSS Statistics version 26.0 (IBM Corp, Armonk, New York, United States). Continuous variables were reported by mean ± standard deviation. Percentages were calculated for categorical data. Chi-square or Fisher's exact tests were used for the distribution of categorical data. Diagnostic accuracy was dichotomized, and binary logistic regression was performed for lesion size, location, and different biopsy needles. Statistical significance was defined as p < 0.05.

Results

There were 111 stereotactic brain biopsy procedures retrieved from our hospital database. One procedure with intraoperative ultrasound and one frame-based biopsy were excluded. In a 7.0-mm pineal region lesion, the probe view suggested that the planned trajectory was close to the internal cerebral vein (3.1 mm) and thalamus (2.4 mm) and we decided on a frame-based biopsy for this case. A total of 106 patients with 109 procedures were analyzed, including 2 patients with two biopsies at two different sites in one operating session and 1 patient with a second biopsy performed for a previous negative biopsy. There were 54 male patients (50.9%) and 52 female patients (49.1%). The mean age at operation was 49.1 ± 19.7 (range: 4.0–80.0) years. The mean operation time was 140 ± 45 (range: 45–250) minutes and the mean operation time of the repeated biopsy was 135 minutes. The locations of the lesions were in the cerebrum in 59 cases (54.1%), insula/basal ganglia/thalamus in 34 cases (31.2%), pinealis/brainstem in 13 cases (11.9%), and cerebellum in 3 cases (2.8%). Eighteen (16.5%) target lesions were less than 10 mm in diameter, while 91 (83.5%) target lesions were ≥10 mm in diameter, as shown in [Table 3]. The 2.5-mm-diameter biopsy needle was used for 30 biopsies (27.5%) from 2007 to 2009 and later the 1.8-mm-diameter needle was used for the other 79 biopsies (72.5%). The BrainLAB navigation system was used in 105 procedures (96.3%) and the Fiagon system was used in 4 procedures (3.7%).

Table 3 The distribution of biopsy cases by lesion location and size

Location	Size of the lesion				Total	Percentage
	≥10 mm	%	< 10 mm	%
Lobar	47	43.1	12	11.0	59	54.1
Insular, basal ganglia, or thalamus	33	30.3	1	0.9	34	31.2
Pinealis, brainstem	8	7.3	5	4.6	13	11.9
Cerebellum	3	2.8	0	0.0	3	2.8
Total	91	83.5	18	16.5	109	100.0

Of the 109 procedures, the diagnostis was conclusive in 103 procedures (94.5%), inconclusive in 4 procedures (3.7%), and negative in 2 procedures (1.8%). Among the inconclusive diagnosis were three cases of brain metastasis and one glioma case. The diagnostic accuracy of each pathology is shown in [Table 4]. Considering the location of the lesion, all (16/16) cerebellar, pineal, and brainstem lesion biopsies were conclusive, 91.2% (31/34) of the biopsies of lesions in the insula, basal ganglia, and thalamus were conclusive, and 94.9% (56/59) of the biopsies of lobar lesions were conclusive. The conclusive diagnosis was not associated with a deep or superficial location of the lesion (p = 0.248), the size of the lesion (≥1 or <1 cm, p = 0.992), or the size of the biopsy needle (2.5 or 1.8 mm, p = 0.744).

Table 4 The diagnostic accuracy for different histologic categories and subtypes

Category			Diagnostic status
		No. of cases (%)	Conclusive	Inconclusive	Negative	Percentage of conclusive diagnosis
Neoplasm	Glioma	51 (46.8)	50 (45.9%)	1 (0.9%)	0 (0.0%)	98.0
	Primary CNS lymphoma	31 (28.4)	30 (27.5%)	0 (0.0%)	1 (0.9%)	96.8
	Other primary brain tumors	8 (7.3)	7 (6.4%)	0 (0.0%)	1 (0.9%)	87.5
	Brain metastasis	6 (5.5)	3 (2.8%)	3 (2.8%)	0 (0.0%)	50.0
Infection	Brain abscess	4 (3.7)	4 (3.7%)	0 (0.0%)	0 (0.0%)	100.0
	Encephalitis	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
	Tuberculosis	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
	Cysticercosis	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
Others	Progressive multifocal leukoencephalopathy	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
	Autoimmune disease	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
	Demyelination	3 (2.8)	3 (2.8%)	0 (0.0%)	0 (0.0%)	100.0
	Radiation necrosis	1 (0.9)	1 (0.9%)	0 (0.0%)	0 (0.0%)	100.0
Total		109 (100)	103 (94.5%)	4 (3.7%)	2 (1.8%)	94.5

Abbreviations: CNS, central nervous system.

The details of the four inconclusive and two negative results were studied ([Table 5]). Among the six biopsies without conclusive results were three cases of inaccurate targeting where the biopsy site deviated from the planned target with reference to the postoperative CT scans. In these three cases, the lesions were in the parieto-occipital region. In one of these three cases, the lesion was small and near the ventricular wall in the trigone. The biopsy needle went into the ventricle and cerebrospinal fluid was aspirated before tissue could be obtained. In another two cases, the biopsies were taken at the nonrepresentative part of the lesions, including the nonenhancing part in the center of one lesion, and the cystic part of the other lesion. In the remaining case, the biopsy was taken after steroid treatment was started, and a subsequent stereotactic biopsy after steroid discontinuation confirmed primary central nervous system (CNS) lymphoma. We evaluated that the planned target site was contained inside the actual biopsy area in 106 (97.2%) procedures.

Table 5 Details of the inconclusive diagnosis and negative biopsy cases with respective error types

Case	Year	Age (y)	Lesion location	Biopsy pathology	Diagnosis	Final diagnosis	Final diagnosis by	Biopsy site	Target selection	Target lesion size (mm)	Error type
1	2007	52	Basal ganglia	Macrophage and chronic inflammation	Negative	Lymphoma	Second biopsy after withholding steroid	Accurate	Appropriate	32	Timing of surgery
2	2008	61	Lobar, trigone	Gliosis	Inconclusive	Brain metastasis	Lung FNAC: non–small cell carcinoma Intracranial cyst fluid aspiration: carcinoma cell present	Deviated	Appropriate	25	Inaccurate targeting
3	2013	54	Lobar, trigon	Necrotic tissue with a small number of highly atypical cells and endothelial proliferation, suggestive of a tumor	Inconclusive	Brain metastasis	Craniotomy, biopsy and excision	Deviated	Appropriate	26	Inaccurate targeting
4	2016	63	Insular	Necrotic tissue	Inconclusive	Brain metastasis	Craniotomy, cyst drainage and biopsy	Accurate	Nonrepresentative	45	Target selection
5	2016	12	Lobar, trigon	blood clot and crushed brain tissue only	Negative	Germinoma	Based on the history and the response to subsequent radiation therapy	Deviated	Appropriate	7	Inaccurate targeting
6	2017	62	Insular	Cyst fluid without tumor cell Tissue: gliosis	Inconclusive	Pilocytic astrocytoma	Craniotomy, biopsy and excision	Accurate	Nonrepresentative	46	Target selection

Abbreviation: FNAC, fine-needle aspiration cytology.

Postoperative hemorrhage was noted in 50 cases (45.9%), with a mean diameter of 9.07 ± 5.32 mm, and a range from 3.00 to 24.00 mm. Postoperative hemorrhage with a size greater than 10 mm occurred in 18 cases (16.5%). The postoperative hemorrhage rates of different biopsy needles are shown in [Table 6]. Primary CNS lymphoma has the highest postoperative hemorrhage rate (64.5%), compared to glioma (39.2%), brain metastasis (33.3%), and other primary brain tumors (37.5%; p = 0.124). Symptomatic hemorrhage was noted in one patient (0.9%) with the 2.5-mm-diameter biopsy needle, who developed right hemiparesis after a biopsy of the left basal ganglia lesion with muscle strength reduction from grade 5 to 4 by the Medical Research Council Scale and a postoperative hematoma of 11 mm in diameter.[8] The largest hemorrhage (24 mm) occurred in a biopsy for a right temporal lesion, with subsequent bleeding into the right atrium. The patient remained asymptomatic, and no intervention was necessary.

Table 6 Comparison of diagnostic accuracy and hemorrhage risk by different biopsy needles

Stereotactic biopsy		Case number (%)			p value
		Total number	2.5-mm-diameter needle	1.8-mm-diameter needle
Total number		109 (100.0%)	30 (27.5%)	79 (17.4%)
Diagnostic accuracy	Conclusive	103 (94.5%)	28 (25.7%)	75 (68.8%)	0.769
	Inconclusive	4 (3.7%)	1 (0.9%)	3 (2.8%)
	Negative	2 (1.8%)	1 (0.9%)	1 (0.9%)
Hemorrhage	Asymptomatic	49 (45.0%)	20 (18.3%)	29 (26.6%)	0.507
	Symptomatic	1 (0.9%)	1 (0.9%)	0 (0.0%)
Size of hemorrhage	< 5 mm	15 (13.8%)	3 (2.8%)	12 (11.0%)	0.104
	5–10 mm	17 (15.6%)	8 (7.3%)	9 (8.3%)
	> 10 mm	18 (16.5%)	10 (9.2%)	8 (7.3%)

Among the 109 cases, 11 cases received craniotomy after biopsy. In 9 cases (82%), the histologic diagnosis was in agreement with the diagnosis from the biopsy. The remaining two cases were nondiagnostic from stereotactic biopsies ([Table 5], cases 1 and 4) and subsequent open biopsies provided the final diagnoses. After the publication of the WHO 2016 classification of CNS tumors, we retrospectively identified that the median number of tissue blocks taken in each biopsy procedure increased from four blocks to five blocks (p = 0.045). IHC staining and molecular studies were accomplished except for three cases of glioblastoma multiforme (GBM), due to insufficient tissue.

A literature review process is illustrated in [Fig. 1], and we have identified 14 original studies evaluating the diagnostic accuracy or diagnostic yield of frameless stereotactic brain biopsy with more than 100 procedures performed from 2011 to 2022 ([Table 7]).[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

Table 7 Comparison of the results from different publications on frameless stereotactic brain biopsy

Study	Center	Year	No. of cases	Navigation system	Diagnostic accuracy (%)	Symptomatic hemorrhage (%)	Mortality (%)
Kubovsky et al[9]	Israel	2022	226	Medtronic StealthStation (Medtronic, Minneapolis, US)	94.2	N/A	N
Mallereau et al[10]	France	2022	212	BrainLAB (BrainLAB AG, Feldkirchen, Germany)	93.3	4.7	0.4
Current study	Hong Kong	2021	109	BrainLAB, Fiagon (Fiagon GmbH, Berlin, Germany)	94.5	0.9	0.0
Ungar et al[11]	Israel	2022	130	Medtronic StealthStation	96.9	1.6	0.8
Giamouriadis et al[12]	United Kingdom	2019	371	Medtronic StealthStation AxiEM	94.1	0.9	1.1
Æbelø et al[22]	Denmark	2019	111	Medtronic Stealthstation	98.2	3.6	1.8
Sciortino et al[13]	Italy	2019	140	BrainLAB	93.6	2.1	0.0
Mader et al[14]	Germany	2019	119	BrainLAB	92.4	1.7	0.8
Kellermann et al[15]	Germany	2017	229	Stryker STP3 (Stryker, Kalamazoo, US)	91.7	0.4	0.4
Verploegh et al[16]	Netherlands	2015	247	Medtronic StealthStation, BrainLAB	94.6	1.2	0.8
Lu et al[17]	United States	2015	113	GE Healthcare	89.4	N/A	N/A
Khatab et al[18]	Netherlands	2014	235	Medtronic StealthStation, Zeiss-MK M (Carl-Zeiss, Oberkochen, Germany)	72.8	2.1	0.9
Air et al[19]	United States	2012	284	Medtronic StealthStation, BrainLAB	89.2	N/A	N/A
Harrisson et al[20]	United Kingdom	2012	150	Medtronic StealthStation AxiEM	96.7	4.7	2.7
Frati et al[21]	Italy	2011	296	Medtronic StealthStation	99.7	1.0	N/A

Abbreviations: N/A, not available.

Fig. 1 The flowchart of the literature review process for the diagnostic accuracy of frameless stereotactic brain biopsy.

Discussion

This study was preceded by our previous publication in 2014 with a new emphasis on diagnostic accuracy and an aim for improvement. We previously reported 92.6% (50/54) accuracy and a symptomatic hemorrhage rate of 1.9% (1/54) in a series of 54 frameless stereotactic brain biopsies.[3] In our current study, the diagnostic accuracy was 94.5% (103/109) for the 109 consecutive procedures using clearly defined criteria. The result was comparable to the previous studies.[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] There were different definitions for diagnostic accuracy or diagnostic yield in the literature, creating heterogenetic results. Æbelø et al[22] and Bradac et al[23] considered diagnostic yield as “the likelihood that a test or procedure will provide the information needed to establish a diagnosis.” Æbelø et al reported a high diagnostic yield of 98.2% for the frameless stereotactic biopsy, but there was 18.9% pathology showing nonspecific reactive change, inflammation, necrosis, no signs of malignancy, gliosis, or normal brain tissue.[22] It creates a dilemma if a biopsy result of “negative for malignancy” is either considered a positive yield (providing information and excluding neoplasm) or a negative yield (not making the true diagnosis). Lu et al[17] and Sciortino et al[13] considered diagnostic yield as the percentage of the number of definitive diagnoses over the total number of cases although there were differences in defining a definitive diagnosis. For example, the biopsy result of cerebral infarct was considered definitive in Lu et al[17] but not in Sciortino et al.[13] The diagnostic accuracy in their studies was 89.4 and 93.6%, respectively. The criteria published by Khatab et al in 2014 improved the certainty and precision of diagnostic yield with stratification for neoplastic cases, nonneoplastic cases, or cases with reactive changes only.[18] Our definition, as previously published by Yuen et al in 2014, was also established to reduce the surgeon's bias in committing whether the histologic diagnosis is accurate or not, and to improve the definition of diagnostic accuracy through multidisciplinary evaluations.[3]

In our case series, there were four inconclusive diagnoses and two negative biopsies. We further analyzed all six procedures in detail with reference to target selection, trajectory planning, intraoperative navigation, aspiration technique, and postoperative biopsy tract verification. The identified causes of inconclusive diagnosis or negative biopsy included two inappropriate target selections, three inaccurate targets, and one incorrect timing, as shown in [Table 5]. These errors have been encountered previously and the question to neurosurgeons is how to avoid them.[6] [24] [25]

One inconclusive biopsy in 2007 was due to incorrect timing ([Table 5], case 1). It was a case of primary central nervous system lymphoma (PCNSLP) who received steroids before the biopsy ([Fig. 2]). It was well known that PCNSLP is very sensitive to steroids. For patients with brain lesions undergoing biopsy while on steroids, an updated image is very important. If there is significant shrinkage of the lesion, one important differential diagnosis is lymphoma and in such case, the biopsy should be withheld.[26] We learned the lesson and no similar error occurred in the subsequent 30 cases of suspected CNS lymphoma. Two inconclusive cases occurred in 2016 and 2017 due to errors in target selection ([Table 5], cases 4 and 6). In one procedure, the target was in the central necrotic area and the other was inside the tumor cyst. Retrospectively, the error might have been avoided if we paid more attention to the characteristic of the imaging finding ([Figs. 3] and [4]). There were three nondiagnostic procedures with inaccurate targets noted on the postoperative CT and all these lesions were in the parietal lobe near the trigone. One case is illustrated in [Fig. 5], in which the biopsy needle entered the ventricle and missed the target. In this case, the error could also be attributed to the small lesion size and proximity to the trigone of the lateral ventricle. Birski et al reported an increased nondiagnostic yield (5.9%, 5/85 cases) for intraventricular or periventricular lesions.[27] Our experience is that using facial recognition for registration, special attention should be paid to the accuracy in the parieto-occipital region. The accuracy can be checked by the function of the navigation software and improved by adding registration points to the parieto-occipital scalp during the registration procedure. Firmly fixed skull pins and the refractory array are also important to avoid inaccuracy in targeting. In addition, we hypothesized that during the registration process by facial surface matching, the patient's facial features might have been shifted by gravity in the supine-oblique position while the navigation CT was performed when the patient was in the neutral position. This problem can be alleviated by performing the facial registration while the patient is in the neutral position and the operation table can be tilted after the registration process for the subsequent biopsy procedure. Alternatively, intraoperative stereotactic CT after the patient's position is finalized may also reduce this type of error.

Fig. 2 A 52-year-old man presented with cognitive decline ([Table 5], case 1). He was put on dexamethasone 4 mg four times per day. (a) Computed tomography (CT) at presentation. (b) CT 16 days after steroid treatment. The biopsy was performed on day 17. (c) CT on the same day after the biopsy. The arrow indicates a dot hemorrhage indicating the biopsy site. (d) CT 1 month after the first biopsy with steroid discontinued. The lesion reappeared. (e) CT after the second biopsy. The arrow indicates the small air bubble indicating the biopsy site.

Fig. 3 A 63-year-old woman presented with progressive lethargy and dysarthria ([Table 5], case 4). (a, b) Preoperative magnetic resonance imaging (MRI) T1 with contrast in sagittal and coronal views. The red line indicates the planned biopsy trajectory. (c, d) Postoperative plain computed tomography (CT) co-registered with the preoperative MRI. The air bubble inside the lesion indicated that the biopsy was accurate. During the operation, the frozen section result was necrotic tissue. Further biopsies were performed at 6 mm deep and 8 mm superficial to the planned target point. However, the final pathologic result remained necrotic tissue only. Craniotomy for open biopsy and excision 2 weeks later confirmed the diagnosis to be metastatic carcinoma. Retrospectively, if the biopsy trajectory was from the low temporal lobe toward the thick part of the enhanced rim of the tumor, a diagnostic specimen might have been obtained.

Fig. 4 A 62-year-old woman presented with right-side weakness and slurring of speech ([Table 5], case 6). (a, b) Preoperative magnetic resonance imaging (MRI) T1 with contrast and (c) preoperative computed tomography (CT) in oblique views aligned to the biopsy trajectory. The red line was the planned trajectory targeting the junction between the cyst and the solid components of the lesion. All images were co-registered. (d) Postoperative CT. The cyst of the lesion collapsed after cyst fluid aspiration. However, no diagnostic tissue was obtained. The pathologic diagnosis was pilocytic astrocytoma on subsequent open biopsy and tumor excision 2 years later. Retrospectively, after cyst fluid aspiration, we might be able to obtain the diagnostic tissue by rotating the window of the biopsy needle inferiorly toward the solid part of the lesion. In addition, intraoperative imaging would be helpful for this case.

Fig. 5 A 12-year-old girl with a known history of germinoma received chemotherapy ([Table 5], case 5). A new lesion was found on surveillance magnetic resonance imaging (MRI) with suspected tumor spread. The lesion was small and located in the parietal lobe near the trigone of the lateral ventricle. (a-d) Preoperative biopsy plan on MRI three-dimensional (3D) reconstruction (red line: trajectory). (a) The position of the patient for facial matching registration during the operation. (e–h) Postoperative computed tomography (CT) was fused to preoperative imaging, which showed that the biopsy tract medially deviated from the planned trajectory. Air bubbles were present in the ventricle and were consistent with the cerebrospinal fluid aspiration during the biopsy procedure. The pathologic diagnosis of the biopsy was mainly a blood clot with some crushed brain tissue. The patient received radiotherapy subsequently and the lesion disappeared.

Reviewing the six cases without a diagnosis, we believed that the error could be definitively avoided in the timing of biopsy for lymphoma and could likely be avoided in the cases with inaccurate targeting. For the two cases with target selection error, avoidance remained possible.

Despite the comparable diagnostic yield from frameless biopsy, frame-based stereotaxy still has advantages for lesions that warrant a higher level of target accuracy, that is, an extremely small, deep-seated lesion with close proximity to the vital structures or vessels. Lunsford et al reported that frameless stereotaxy achieved excellent accuracy after meticulous registration, with an error of ± 2 mm.[28] This error may not be acceptable for certain cases and frame-based biopsy was the better alternative.

Euclidean distance is frequently utilized to evaluate the stereotactic accuracy for functional neurosurgical procedures.[29] [30] [31] However, its usage is limited in stereotactic biopsy procedures where the biopsy tract is not always clearly visible. In addition, the surgeon might take several biopsies in different directions or a few millimeters superficial or deeper to the target site to increase the amount of specimen. The actual biopsy site is often represented by an area of hypodense air bubbles or a punctate hemorrhage on the postoperative scan. The planned bony entry point was often obscured by the burr hole made for the biopsy on the postoperative CT. All these factors affected calculations for the deviation distance from the planned biopsy target and the actual biopsy site. Kubovsky et al suggested the presence of air bubbles outside the biopsy target on postoperative CT fused with preoperative MRI warranted a second biopsy.[9] No other publications provided the actual deviation distance.

Apart from a positive histologic diagnosis for tumor type and grading, further IHC staining and molecular evaluations were increasingly important in the diagnosis and prognosis of certain types of brain tumors. Eigenbrod et al reported that in 3.1% of cases (5/159), the final diagnosis was altered by molecular findings.[32] For the GBM cases, the absence of the TERT mutation and the presence of the MGMT promoter methylation carried a better prognosis.[33] [34] Mader et al reported a nondiagnostic rate of 1.5% (1 in 67 cases) for the MGMT promoter methylation status among the GBM cases in 2018.[14] We have three GBM cases where IHC staining was incomplete due to insufficient tumor sampling, including two GBM cases with the TERT mutation detection not performed and one GBM case with IHC staining unsuccessful except for the IDH status. It is conceivable that the volume of tissue sampled in each case and the target selection was also important for the molecular diagnosis. Among the three GBM cases with partial IHC staining, four to eight blocks of the sample were taken, but there was a relatively small amount of tissue with necrosis in each block, which jeopardized the molecular diagnosis. Weise et al suggested that the tangential trajectory planning along the long axis of the lesion would enhance the MGMT promoter methylation detection.[35] Katzendobler et al reported a success rate of 93% for molecular diagnosis by stereotactic biopsy for CNS tumors.[36] Our experience is that four appropriately sized specimens with intraoperative confirmation from the pathologist for the quality and quantity of the specimen are essential for histologic and IHC staining. The development of rapid intraoperative molecular study may further improve the diagnostic accuracy for glioma.[37]

Another important topic for stereotactic brain biopsy is to examine the agreement of pathologic diagnosis between the biopsy and open surgery. The discrepancy between stereotactic biopsy and open surgery had been reported to be 38% in Jackson et al,[38] 24% in Reithmeier et al,[39] and relatively low (3%) in Kim et al.[40] The reason for the discrepancy between stereotactic biopsy and open surgery was that the biopsy site might not fully represent the entire lesion and the true diagnosis might not be reached.[38] Pasternak et al suggested that biopsies at multiple sites are important for the accurate diagnosis of heterogeneous lesions.[41] Among the 11 cases that received both biopsy and open surgery in our case series, there were 2 cases with discrepancies in the biopsy and subsequent excision result and thus the reliability rate was 82%. These two cases were known to be nondiagnostic after the stereotactic biopsy. We believed that appropriate target selection could improve reliability and avoid this discrepancy.

Our previous cohort by Yuen et al discussed the variation from 3.4 to 50.0% in the prevalence of asymptomatic hemorrhage after stereotactic biopsies.[3] Kesserwan et al reported an asymptomatic hemorrhage rate of 15.7% and a symptomatic hemorrhage rate of 3.9% in the systemic review for frameless stereotactic biopsy, while the overall hemorrhage risk of frame-based stereotactic biopsy was reported to be 21.3% (range: 7.0–59.8%).[42] [43] A large series of 1,500 consecutive frame-based stereotactic biopsies suggested an asymptomatic hemorrhage rate of 22.4%.[44] Mathon et al[45] reported a 32.2% postoperative hemorrhage rate and 4.5% symptomatic complication rate in 177 patients with primary CNS lymphoma in the subset of data from Riche et al.[44] The asymptomatic hemorrhage rate for primary CNS lymphoma in our series was 64.5% (20/31 cases) and no patient developed symptomatic hemorrhage. An example of a punctate hemorrhage after stereotactic biopsy for a primary CNS lymphoma is illustrated in [Fig. 6].

Fig. 6 A 52-year-old man with a left frontal tumor underwent a frameless stereotactic biopsy. Postoperative computed tomography (CT) showed a small punctate hemorrhage of 4.2 mm. The final pathology was lymphoma and the patient did not have any new neurologic deficit after the biopsy. (a, b) A small punctate hemorrhage at the biopsy site on raw and labeled axial images. (c) An air bubble demonstrating the biopsy site (white arrow).

A higher rate of asymptomatic hemorrhage rate was attributed to the inclusion of the punctate hemorrhage and the postoperative assessment by a fine-cut CT. There is certain heterogeneity in the definitions of postbiopsy hemorrhage and punctate hemorrhage was not included in Grossman et al[46] and Shakal et al,[47] in which low postbiopsy hemorrhage rates of 7% and 4.7% were reported, respectively. The prospective study by Kulkarni et al revealed an overall hemorrhage rate of 59.8% for patients who underwent frame-based brain biopsy with 22.5% of hemorrhage <5 mm, 20.6% of hemorrhage between 5 and 10 mm, a 50% asymptomatic hemorrhage rate, and 4.9% symptomatic hemorrhage rate.

The presence and size of hemorrhage on postoperative CT depends on the duration from operation to imaging and the slice interval of CT. Using conventional CT with a 5-mm slice thickness, a small hemorrhage of less than 5 mm could be concealed and the hemorrhage size could be underestimated for 5 mm. We revealed all types of postoperative hemorrhage, including silent or punctate hemorrhage, asymptomatic hemorrhage, or symptomatic hemorrhage, and measured the hemorrhage size with fine-cut CT of 0.625-mm slice interval. The symptomatic hemorrhage rate was only 0.9% and remained low in comparison with the other published literature ([Table 7]). Our study enhanced the belief that a 1.8-mm-diameter biopsy needle provided a lower incidence of symptomatic hemorrhage without compromising the diagnostic accuracy.

Limitations

There were certain limitations in our research. First, this was a retrospective study with an extended study interval, and experience accumulation might weaken the comparability between early and recent results. Second, there was heterogeneity in the number of aspirations taken for each biopsy. Third, the annual number of biopsy cases was relatively low. However, the annual number of primary intracranial tumors was estimated to be 8 to 9 per 100,000 population per year in our locality and we estimated that 100 to 110 cases of intracranial tumors were treated each year by open surgery or biopsy.[48] Despite these limitations, our analysis of the diagnostic accuracy and illustration of the nondiagnostic cases provided insights for further improvements in this surgical technique.

Conclusion

Frameless stereotactic biopsy is an accurate procedure with a high diagnostic accuracy rate and a low symptomatic hemorrhage rate. Deviations may occur for lesions in the parieto-occipital or trigonal region, and biopsy at the nonrepresentative site of the lesion introduces an inconclusive diagnosis or a negative biopsy. Periodic review and vigorous auditing of each biopsy result are indispensable in order to avoid pitfalls, evaluate the errors, and ultimately improve surgical outcomes.

Conflict of Interest

None declared.

Disclaimer

The study result was presented in the oral format at the Annual Scientific Meeting of the Hong Kong Neurosurgical Society on December 18, 2020, by Dr. He Zhexi and in video format at the 7th World Federation of Neurosurgical Societies (WFNS) Foundation Asian Congress of Neurological Surgeons (ACNS) Winter Web Seminar on February 20, 2021, by Dr. He Zhexi.

A preprint version of the older version of this manuscript is available on Research Square with DOI:10.21203/rs.3.rs-854266/v1 and was cited in the manuscript as Reference 4. The current manuscript included the latest publications in the literature review and updated the discussion.

Availability of Data and Material

The data that support the findings of this study are available upon request from the corresponding author, under the regulations of the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee.

Ethical Approval

This research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. The study protocol was reviewed and approved by the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee (CREC Ref. No. 2021.354).

Consent to Participate

The study protocol is reviewed and approved by the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee (CREC Ref. No. 2021.354) with the patient consent waived as this research is based on secondary analysis of available data with no additional risk to patients. There is no patient-identifiable information contained in this manuscript.

• References

• 1 Dhawan S, He Y, Bartek Jr J, Alattar AA, Chen CC. Comparison of frame-based versus frameless intracranial stereotactic biopsy: systematic review and meta-analysis. World Neurosurg 2019; 127: 607-616.e4
• 2 Woodworth GF, McGirt MJ, Samdani A, Garonzik I, Olivi A, Weingart JD. Frameless image-guided stereotactic brain biopsy procedure: diagnostic yield, surgical morbidity, and comparison with the frame-based technique. J Neurosurg 2006; 104 (02) 233-237
• 3 Yuen J, Zhu CX, Chan DT. et al. A sequential comparison on the risk of haemorrhage with different sizes of biopsy needles for stereotactic brain biopsy. Stereotact Funct Neurosurg 2014; 92 (03) 160-169
• 4 He Z, Zhu XL, Chan TMD. et al. The diagnostic accuracy and era for improvement for frameless stereotactic brain biopsy: a series of 109 procedures and a systemic review. 2021; (e-pub ahead of print) DOI: 10.21203/rs.3.rs-854266/v1.
• 5 Agha RA, Fowler AJ, Rajmohan S, Barai I, Orgill DP. PROCESS Group. Preferred reporting of case series in surgery; the PROCESS guidelines. Int J Surg 2016; 36 (Pt A): 319-323
• 6 Krieger MD, Chandrasoma PT, Zee CS, Apuzzo ML. Role of stereotactic biopsy in the diagnosis and management of brain tumors. Semin Surg Oncol 1998; 14 (01) 13-25
• 7 Ringel F, Ingerl D, Ott S, Meyer B. VarioGuide: a new frameless image-guided stereotactic system: accuracy study and clinical assessment. Neurosurgery 2009; 64(5, Suppl 2):365–371, discussion 371–373
• 8 Medical Research Council. Aids to Examination of the Peripheral Nervous System. Memorandum No. 45. Superseding War Memorandum No. 7. London: Her Majesty's Stationery Office; 1976
• 9 Kubovsky S, Khriesh A, Moscovici S, Paldor I. Fusion of preoperative and postoperative imaging may predict the diagnostic yield of stereotactic needle brain biopsies. World Neurosurg 2022; 157: e441-e447
• 10 Mallereau CH, Chibbaro S, Ganau M. et al. Pushing the boundaries of accuracy and reliability during stereotactic procedures: a prospective study on 526 biopsies comparing the frameless robotic and image-guided surgery systems. J Clin Neurosci 2022; 95: 203-212
• 11 Ungar L, Nachum O, Zibly Z. et al. Comparison of frame-based versus frameless image-guided intracranial stereotactic brain biopsy: a retrospective analysis of safety and efficacy. World Neurosurg 2022; 164: e1-e7
• 12 Giamouriadis A, Perera D, Safdar A. et al. Safety and accuracy of frameless electromagnetic-navigated (AXIEM™)-guided brain lesion biopsies: a large single-unit study. Acta Neurochir (Wien) 2019; 161 (12) 2587-2593
• 13 Sciortino T, Fernandes B, Conti Nibali M. et al. Frameless stereotactic biopsy for precision neurosurgery: diagnostic value, safety, and accuracy. Acta Neurochir (Wien) 2019; 161 (05) 967-974
• 14 Mader MM, Rotermund R, Martens T, Westphal M, Matschke J, Abboud T. The role of frameless stereotactic biopsy in contemporary neuro-oncology: molecular specifications and diagnostic yield in biopsied glioma patients. J Neurooncol 2019; 141 (01) 183-194
• 15 Kellermann SG, Hamisch CA, Rueß D. et al. Stereotactic biopsy in elderly patients: risk assessment and impact on treatment decision. J Neurooncol 2017; 134 (02) 303-307
• 16 Verploegh IS, Volovici V, Haitsma IK. et al. Contemporary frameless intracranial biopsy techniques: might variation in safety and efficacy be expected?. Acta Neurochir (Wien) 2015; 157 (11) 2011-2016 , discussion 2016
• 17 Lu Y, Yeung C, Radmanesh A, Wiemann R, Black PM, Golby AJ. Comparative effectiveness of frame-based, frameless, and intraoperative magnetic resonance imaging-guided brain biopsy techniques. World Neurosurg 2015; 83 (03) 261-268
• 18 Khatab S, Spliet W, Woerdeman PA. Frameless image-guided stereotactic brain biopsies: emphasis on diagnostic yield. Acta Neurochir (Wien) 2014; 156 (08) 1441-1450
• 19 Air EL, Warnick RE, McPherson CM. Management strategies after nondiagnostic results with frameless stereotactic needle biopsy: retrospective review of 28 patients. Surg Neurol Int 2012; 3 (Suppl. 04) S315-S319
• 20 Harrisson SE, Shooman D, Grundy PL. A prospective study of the safety and efficacy of frameless, pinless electromagnetic image-guided biopsy of cerebral lesions. Neurosurgery 2012; 70(1, Suppl Operative):29–33, discussion 33
• 21 Frati A, Pichierri A, Bastianello S. et al. Frameless stereotactic cerebral biopsy: our experience in 296 cases. Stereotact Funct Neurosurg 2011; 89 (04) 234-245
• 22 Æbelø AM, Noer VR, Schulz MK, Kristensen BW, Pedersen CB, Poulsen FR. Frameless stereotactic neuronavigated biopsy: a retrospective study of morbidity, diagnostic yield, and the potential of fluorescence—a single-center clinical investigation. Clin Neurol Neurosurg 2019; 181: 28-32
• 23 Bradac O, Steklacova A, Nebrenska K, Vrana J, de Lacy P, Benes V. Accuracy of VarioGuide frameless stereotactic system against frame-based stereotaxy: prospective, randomized, single-center study. World Neurosurg 2017; 104: 831-840
• 24 Mansour MH. Incidence of miss targeting in frame-based stereotactic brain surgery. Egypt J Hosp Med 2018; 73 (09) 7454-7457
• 25 Giannini C, Dogan A, Salomão DR. CNS lymphoma: a practical diagnostic approach. J Neuropathol Exp Neurol 2014; 73 (06) 478-494
• 26 Deckert M, Engert A, Brück W. et al. Modern concepts in the biology, diagnosis, differential diagnosis and treatment of primary central nervous system lymphoma. Leukemia 2011; 25 (12) 1797-1807
• 27 Birski M, Furtak J, Krystkiewicz K. et al. Endoscopic versus stereotactic biopsies of intracranial lesions involving the ventricles. Neurosurg Rev 2021; 44 (03) 1721-1727
• 28 Lunsford LD, Niranjan A, Khan AA, Kondziolka D. Establishing a benchmark for complications using frame-based stereotactic surgery. Stereotact Funct Neurosurg 2008; 86 (05) 278-287
• 29 Lee J, Huang Z, Lee S. Accurate stereotaxic localization using computerized tomography with geometric correction. Biomed Eng Appl Basis Commun 2002; 14 (05) 189-196
• 30 Atsumi H, Matsumae M. Fusing of preoperative magnetic resonance and intraoperative O-arm images in deep brain stimulation enhance intuitive surgical planning and increase accuracy of lead placement. Neurol Med Chir (Tokyo) 2021; 61 (05) 341-346
• 31 Sharma M, Rhiew R, Deogaonkar M, Rezai A, Boulis N. Accuracy and precision of targeting using frameless stereotactic system in deep brain stimulator implantation surgery. Neurol India 2014; 62 (05) 503-509
• 32 Eigenbrod S, Trabold R, Brucker D. et al. Molecular stereotactic biopsy technique improves diagnostic accuracy and enables personalized treatment strategies in glioma patients. Acta Neurochir (Wien) 2014; 156 (08) 1427-1440
• 33 Eckel-Passow JE, Lachance DH, Molinaro AM. et al. Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med 2015; 372 (26) 2499-2508
• 34 Hegi ME, Diserens AC, Gorlia T. et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352 (10) 997-1003
• 35 Weise LM, Harter PN, Eibach S. et al. Confounding factors in diagnostics of MGMT promoter methylation status in glioblastomas in stereotactic biopsies. Stereotact Funct Neurosurg 2014; 92 (03) 129-139
• 36 Katzendobler S, Do A, Weller J. et al. Diagnostic yield and complication rate of stereotactic biopsies in precision medicine of gliomas. Front Neurol 2022; 13: 822362
• 37 Shankar GM, Francis JM, Rinne ML. et al. Rapid intraoperative molecular characterization of glioma. JAMA Oncol 2015; 1 (05) 662-667
• 38 Jackson RJ, Fuller GN, Abi-Said D. et al. Limitations of stereotactic biopsy in the initial management of gliomas. Neuro-oncol 2001; 3 (03) 193-200
• 39 Reithmeier T, Lopez WO, Doostkam S. et al. Intraindividual comparison of histopathological diagnosis obtained by stereotactic serial biopsy to open surgical resection specimen in patients with intracranial tumours. Clin Neurol Neurosurg 2013; 115 (10) 1955-1960
• 40 Kim JE, Kim DG, Paek SH, Jung HW. Stereotactic biopsy for intracranial lesions: reliability and its impact on the planning of treatment. Acta Neurochir (Wien) 2003; 145 (07) 547-554 , discussion 554–555
• 41 Pasternak KA, Schwake M, Warneke N. et al. Evaluation of 311 contemporary cases of stereotactic biopsies in patients with neoplastic and non-neoplastic lesions-diagnostic yield and management of non-diagnostic cases. Neurosurg Rev 2021; 44 (05) 2597-2609
• 42 Riche M, Amelot A, Peyre M, Capelle L, Carpentier A, Mathon B. Complications after frame-based stereotactic brain biopsy: a systematic review. Neurosurg Rev 2021; 44 (01) 301-307
• 43 Kesserwan MA, Shakil H, Lannon M. et al. Frame-based versus frameless stereotactic brain biopsies: a systematic review and meta-analysis. Surg Neurol Int 2021; 12: 52
• 44 Riche M, Marijon P, Amelot A. et al. Severity, timeline, and management of complications after stereotactic brain biopsy. J Neurosurg 2021; 136 (03) 867-876
• 45 Mathon B, Marijon P, Riche M, Amelot A. PSL Brain-Biopsy Study Group. Letter to the Editor regarding “Hemorrhagic attitude in frameless and frame-based stereotactic biopsy for deep-seated primary central nervous system lymphomas in immunocompetent patients: a multicentric analysis of the last twenty years.”. World Neurosurg 2021; 152: 242-243
• 46 Grossman R, Sadetzki S, Spiegelmann R, Ram Z. Haemorrhagic complications and the incidence of asymptomatic bleeding associated with stereotactic brain biopsies. Acta Neurochir (Wien) 2005; 147 (06) 627-631 , discussion 631
• 47 Shakal AAS, Mokbel EAH. Hemorrhage after stereotactic biopsy from intra-axial brain lesions: incidence and avoidance. J Neurol Surg A Cent Eur Neurosurg 2014; 75 (03) 177-182
• 48 He Z, Wong ST, Yam KY. Newly-diagnosed, histologically-confirmed central nervous system tumours in a regional hospital in Hong Kong : an epidemiological study of a 21-year period. J Korean Neurosurg Soc 2020; 63 (01) 119-135

Publikationsverlauf

Eingereicht: 20. Juni 2022

Angenommen: 05. Dezember 2022

Accepted Manuscript online:
08. Dezember 2022

Artikel online veröffentlicht:
11. Mai 2023