Intracranial hemorrhagic risk after cortical brain biopsy during shunt surgery for idiopathic normal pressure hydrocephalus: a prospective comparative study

Abstract

Background

This study prospectively compared postoperative intracranial hemorrhagic complications following shunt surgery for idiopathic normal pressure hydrocephalus (iNPH) in patients undergoing the procedure with or without a concomitant cortical brain biopsy.

Methods

All patients undergoing shunt surgery for iNPH at our institution between January 1, 2018, and December 31, 2025, were prospectively enrolled. Patients were allocated to shunt surgery with or without a cortical brain biopsy. Postoperative intracranial complications were assessed using cranial computed tomography (CT), with predefined outcomes including (1) symptomatic intracranial hemorrhage and (2) asymptomatic minor radiological hemorrhage or infarction. In addition, intracranial hemorrhages requiring surgical treatment within three months postoperatively were recorded.

Results

A total of 588 patients was included (294 per group). Symptomatic intracranial hemorrhage occurred in 2/588 patients (0.3%), with one case in each group (risk difference 0.0%, 95% CI − 0.9 to + 0.9). Asymptomatic radiological findings were observed in 6.3% of patients and were similarly distributed between groups (risk difference − 0.3%, 95% CI − 4.3 to + 3.6). Intracranial hemorrhage requiring surgical treatment within 3 months occurred in 3.6% of patients, without group differences.

Conclusion

In this large prospective cohort, intracranial hemorrhagic complications after shunt surgery for iNPH were rare. No signal of increased hemorrhagic risk was observed with the addition of cortical brain biopsy although small differences in rare complications cannot be excluded.

Introduction

Idiopathic normal pressure hydrocephalus (iNPH) is a potentially reversible cause of gait disturbance, cognitive impairment, and urinary incontinence, and shunt surgery remains the only established effective treatment [14]. While traditionally regarded as a disorder of cerebrospinal fluid (CSF) dynamics, increasing evidence suggests that iNPH also involves neurodegenerative processes [8, 12]. Accordingly, iNPH may be conceptualized as a combined CSF disorder and proteinopathy [2].

Neuropathological studies have demonstrated overlap between iNPH and Alzheimer’s disease (AD), with cortical accumulation of amyloid-β (Aβ) and tau observed in a substantial proportion of patients. Cortical brain biopsies obtained during shunt surgery have shown amyloid deposition in approximately 40% and tau pathology in about 10% of iNPH patients [12]. These protein aggregates are hallmark features of AD and iNPH has been proposed as a model of prodromal AD [13].

Detection of amyloid and tau pathology in cortical biopsies may have important implications for individualized management. A recent systematic review concluded that cortical tissue analysis of Aβ aggregation provides clinically useful information [8]. Several studies have linked amyloid and tau pathology to increased risk of progression to Alzheimer disease [12, 15]. Amyloid pathology in cortical biopsies [1, 9, 19] or elevated brain tissue Aβ42 levels [1] has been associated with reduced response to shunt surgery. Thus, in patients who fail to improve after shunting, identification of concomitant neurodegenerative pathology may guide further evaluation and care.

Despite increasing emphasis on biomarker-based diagnosis in AD, including CSF, blood, and imaging biomarkers, such stratified approaches are not yet established in iNPH. Because cortical brain biopsy can be performed during routine shunt surgery without additional procedures, it represents a potentially accessible method for obtaining disease-relevant information.

However, relatively few centers have adopted cortical brain biopsy during shunt surgery [7, 12, 19], and concerns regarding procedural safety persist. In particular, the risk of intracranial hemorrhage or cerebral infarction associated with the biopsy has not been systematically evaluated. To date, no large prospective comparative study has assessed whether adding a cortical brain biopsy increases postoperative intracranial hemorrhage complications in iNPH. Defining the safety profile of cortical brain biopsy is therefore essential for its potential integration into routine iNPH surgery.

Therefore, we conducted a prospective and comparative safety study including all iNPH patients undergoing shunt surgery at our institution over an eight-year period. The diagnostic utility of a brain biopsy was beyond the scope of this study; instead, we specifically evaluated the safety of cortical brain biopsy by addressing the following questions: (1) What is the incidence of symptomatic intracranial hemorrhage and asymptomatic radiological hemorrhage or infarction following shunt surgery for iNPH? (2) Does the occurrence of these complications differ between patients undergoing shunt surgery with or without a concomitant cortical brain biopsy?

Materials and methods

Study approval and design

Beginning January 1, 2018, all patients undergoing shunt surgery iNPH at our institution were prospectively registered in a dedicated iNPH shunt registry, with systematic recording of perioperative complications and postoperative outcomes.

The study design is summarized in Fig. 1. Allocation was performed pragmatically (non-randomized) based on surgical team availability to shunt surgery either with or without a simultaneous cortical brain biopsy. This allocation strategy reflects routine clinical practice but introduces potential selection bias. However, as allocation was determined by surgical team availability rather than patient characteristics, systematic clinical selection bias was likely limited.

Fig. 1.

Overview of the study design. The study included prospective enrollment and allocation of patients with idiopathic normal pressure hydrocephalus to shunt surgery with or without concomitant cortical brain biopsy

Patient population

The study included all patients diagnosed with iNPH and treated with shunt surgery between January 1, 2018, and December 31, 2025. Diagnosis of iNPH was established according to institutional criteria based on clinical presentation, neuroimaging, and supplementary diagnostic testing [4, 5]. In general, we offer diagnostic assessment and consideration for surgery to patients referred with possible or probable iNPH in accordance with the American-European guidelines [21]. No patients were excluded based on age or sex.

Surgical procedure

Shunt surgery was performed according to standard institutional practice. Procedures were carried out by neurosurgeons with varying levels of experience, reflecting routine clinical care. Shunt surgeries without brain biopsy (no-biopsy group) followed the conventional departmental protocol. In contrast, shunt surgeries including a cortical brain biopsy (biopsy group) were performed by a dedicated surgical team, with the biopsy procedure carried out by an experienced micro-neurosurgeon (P.K.E.) using the operating microscope.

Independent of group allocation, the surgical approach included a right frontal skin incision, trepanation, dural opening, limited cortectomy, and insertion of a ventricular catheter. In patients allocated to the biopsy group, an additional step was performed prior to catheter insertion: a biopsy needle was introduced through the cortectomy site, and a cylindrical cortical tissue specimen measuring approximately 0.9 mm in diameter and 10 mm in length was aspirated. The specimen included pia mater and superficial cortical tissue. Meticulous hemostasis was ensured following tissue sampling. The biopsy was obtained immediately prior to ventricular catheter insertion and was located at the same cortical entry site used for ventricular catheter insertion.

The remaining steps of the procedure were identical between groups and included subcutaneous tunneling of the shunt system and distal catheter placement either into the peritoneal cavity (ventriculoperitoneal shunt) or the right atrium (ventriculoatrial shunt), according to clinical indications.

Assessment of intracranial complications

The safety aspect of obtaining a brain biopsy was assessed from cranial computed tomography (CT), assessing postoperative intracranial complications. CT imaging was routinely performed either approximately 6 h after surgery or on the first postoperative day (Day 1). All CT scans were evaluated by in-house neuroradiologists. We secondarily included the CT scans performed after 1 week and after 6–12 weeks.

In cases of postoperative neurological symptoms or clinical deterioration, additional CT imaging was obtained prior to patient transfer or discharge. As our institution serves as the sole neurosurgical referral center for iNPH patients in the region, patients experiencing postoperative complications are routinely referred back for evaluation.

Postoperative outcomes assessed in this study were predefined as:

• Symptomatic intracranial hemorrhage, defined as radiologically verified intracranial hemorrhage associated with neurological symptoms.

• Asymptomatic minor radiological findings, including small intracranial hemorrhages or ischemic lesions detected on CT without accompanying clinical symptoms.

In addition, the occurrence of subdural hematoma within the first three postoperative months was recorded.

Statistical analysis

Given the low event rates, analyses were primarily descriptive, and confidence intervals were used to assess the range of possible differences between groups. Statistical analyses were performed using IBM SPSS Statistics 31 (IBM Corporation, Armonk, NY, USA). Categorical data are presented as absolute numbers with percentages relative to the total cohort or subgroup. Continuous variables are presented as mean ± standard deviation. Differences between categorical variables were analyzed using Pearson’s chi-square test, while differences between continuous variables were analyzed using independent samples t-tests. All tests were two-sided, and p < 0.05 was considered statistically significant.

Results

Patient cohort

During the eight-year study period from January 1, 2018, to December 31, 2025, a total of 588 patients with iNPH underwent shunt surgery at our institution. The cohort included 350 men and 238 women, with a mean age of 73.7 ± 6.6 years (Table 1).

Table 1.

Study cohort and information about surgery

	Total	Biopsy-group	No-biopsy-group	P-value
Demography
N	588	294	294
Sex (M/F)	350/238	180/114	170/124	0.40
Age (years)	73.7 ± 6.6	73.6 ± 6.0	73.8 ± 7.1	0.77
Surgery
Surgery time (min)	47.5 ± 16.3	41.2 ± 10.1	53.9 ± 18.7	< 0.001
Shunt type (VP/VA)	557/31	273/21	284/10	0.04

Values are presented as numbers with percentages in parenthesis or mean ± standard deviation, unless otherwise indicated

Of these, 294 patients were allocated to shunt surgery with a cortical brain biopsy (biopsy group) and 294 to shunt surgery without biopsy (no-biopsy group). The two groups were comparable with respect to age and sex distribution (Table 1).

Surgical characteristics

Mean surgical time was significantly shorter in the biopsy group compared with the no-biopsy group (41.2 ± 10.1 min vs 53.9 ± 18.7 min; p < 0.001; Table 1). This difference likely reflects differences in surgical team experience rather than the biopsy procedure itself.

Ventriculoperitoneal shunts were used in most patients (557/588; 94.7%), while ventriculoatrial shunts were placed in 31/588 patients (5.3%). The distribution of shunt type differed modestly between groups (p = 0.04; Table 2).

Table 2.

Outcome data

	Total (n = 588)	Biopsy-group (n = 294)	No-biopsy-group (n = 294)	P-value
DAY 1
Symptomatic intracranial haemorrhage on CT (> 1.5 cm)	2 (0.3%)	1 (0.3%)	1 (0.3%)	1.0
Intracerebral haemorrhage	1 (0.2%)	0	1 (0.3)	0.32
Subdural haemorrhage	1 (0.2%)	1 (0.3%)	0	0.32
Asymptomatic haemorrhage or infarction on CT	37 (6.3%)	18 (6.1%)	19 (6.5%)	0.87
Intracerebral haemorrhage (diameter ≤ 1.5 cm)	2 (0.3%)	1 (0.3%)	1 (0.3%)	1.0
Minimal CT findings	35 (6.0%)	17 (5.8%)	18 (6.1%)	0.86
Minor blood within occipital horn of cerebral ventricles	15 (2.6%)	8 (2.7%)	7 (2.4%)	0.79
Minor blood in subarachnoid or subdural space	7 (1.2%)	5 (1.7%)	2 (0.7%)	0.25
Minor blood along ventricular catheter	10 (1.7%)	4 (1.4%)	6 (2.0%)	0.52
Small ischemia along ventricular catheter	3 (0.5%)	0	3 (1.0%)	0.08
AFTER 1 WEEK
CT findings requiring surgery	3 (0.5%)
Craniotomy for acute subdural hematoma	1 (0.2%)	0	1 (0.3%)	0.32
Trepanation for chronic subdural hematoma	2 (0.3%)	2 (0.7%)	0	0.16
AFTER 6–12 WEEKS
CT findings requiring surgery
Craniotomy for acute subdural hematoma	0	0	0
Trepanation for chronic subdural hematoma	18 (3.1%)	10 (3.4%)	8 (2.7%)	0.63

Values are presented as numbers with percentages in parenthesis or mean ± standard deviation, unless otherwise indicated. Significant differences between categorical values were determined by Peason Chi-square test, and differences between continuous data were determined by independent samples t-test

Symptomatic intracranial hemorrhage Day 1

No intracranial hemorrhagic complications required immediate surgical intervention during the initial postoperative period.

Detailed outcome data are presented in Table 2, while absolute risk differences with 95% confidence intervals are summarized in Table 3. Symptomatic intracranial hemorrhage > 1.5 cm on postoperative CT (Day 1) occurred in 2 of 588 patients (0.3%; Table 2). One patient in the biopsy group, an 82-year-old woman, developed an acute subdural hematoma (Fig. 2A). The hematoma was initially asymptomatic but progressed to a chronic subdural hematoma requiring trepanation three weeks later. The patient subsequently improved clinically and was classified as a shunt responder.

Table 3.

Absolute risk differences between groups with 95% confidence intervals

Outcome	Biopsy-group (n = 294)	No-biopsy-group (n = 294)	Risk difference (%)	95% CI (%)
Symptomatic intracranial hemorrhage	1 (0.3%)	1 (0.3%)	0.0	− 0.9 to + 0.9
Asymptomatic radiological findings	18 (6.1%)	19 (6.5%)	− 0.3	− 4.3 to + 3.6
Subdural hematoma requiring surgery (≤ 3 months)	12 (4.1%)	9 (3.1%)	+ 1.0	− 2.0 to + 4.0
Any intracranial complication*	19 (6.5%)	20 (6.8%)	− 0.3	− 4.5 to + 3.8

*Includes symptomatic hemorrhage and asymptomatic radiological findings

Risk differences calculated as biopsy group minus no-biopsy group. CI Confidence interval

Fig. 2.

Representative axial postoperative computed tomography (CT) scans from study participants. A Symptomatic acute subdural hematoma in an 82-year-old woman in the biopsy group, initially asymptomatic and treated with trepanation three weeks later after progression to a chronic subdural hematoma. B Symptomatic postoperative intracerebral hematoma in the right frontal lobe of a 64-year-old woman in the no-biopsy group, managed conservatively. C Asymptomatic intracerebral hemorrhage measuring 10 mm in diameter adjacent to the ventricular catheter in a 63-year-old woman in the biopsy group. D Asymptomatic intracerebral hemorrhage measuring 15 mm in diameter adjacent to the ventricular catheter in a 75-year-old man in the no-biopsy group. E Minor intraventricular blood within the occipital horn of the lateral ventricle. F Minor blood within the subarachnoid space adjacent to the ventricular catheter. G Trace amounts of blood along the ventricular catheter tract. (H) Small ischemic lesion adjacent to the ventricular catheter tract

One patient in the no-biopsy group, a 64-year-old woman, developed a postoperative intracerebral hematoma in the right frontal lobe (Fig. 2B), accompanied by reduced level of consciousness (Glasgow Coma Scale score of 11). The hematoma was managed conservatively, and although the patient recovered neurologically, she did not demonstrate a definitive clinical response to shunt surgery.

The incidence of symptomatic intracranial hemorrhage was identical in the biopsy and no-biopsy groups (0.34% vs 0.34%; risk difference 0.0%, 95% CI − 0.94% to + 0.94%; Table 3).

Asymptomatic radiological findings Day 1

Asymptomatic intracranial hemorrhages or ischemic lesions were identified on postoperative CT (Day 1) in 37 of 588 patients (6.3%; Table 2). These findings included small intracerebral hemorrhages (< 1.5 cm in diameter), minor blood deposits within the occipital horn of the lateral ventricles, small amounts of blood within the subarachnoid or subdural spaces, traces of blood along the ventricular catheter, and small ischemic lesions adjacent to the catheter tract (Fig. 2C–H).

The frequency of asymptomatic radiological findings was similar between the biopsy group (18/294; 6.1%) and the no-biopsy group (19/294; 6.5%) (p = 0.87; Table 2). For asymptomatic radiological findings, the risk difference was − 0.3% (95% CI − 4.3% to + 3.6%; Table 3).

The two largest asymptomatic intracerebral hemorrhages measured 10 mm in diameter in a 63-year-old woman in the biopsy group (Fig. 2C) and 15 mm in diameter in a 75-year-old man in the no-biopsy group (Fig. 2D).

Intracranial hemorrhage requiring surgery after 1 and 6–12 weeks

About 1 week postoperatively, two subjects required trepanation for chronic subdural hematoma in the biopsy group and one required craniotomy for acute subdural hematoma in the no-biopsy group (Table 2). After 6–12 weeks the occurrence of intracranial hemorrhages requiring surgery included 10 chronic subdural hematomas in the biopsy group (3.4%) and eight in the no-biopsy group (2.7%) (Table 2). Intracranial hemorrhage requiring surgical treatment within three months occurred in 21 of 588 patients (3.6%), including cases identified both within the first postoperative week and during later follow-up. For subdural hematomas requiring surgical treatment within three months, the risk difference was + 1.0% (95% CI − 2.0% to + 4.0%; Table 3).

Discussion

The principal finding of this study is the absence of any observable signal of increased risk of intracranial hemorrhagic complications when cortical brain biopsy is performed during shunt surgery for iNPH, in a large prospective cohort. This is the first large prospective comparative study specifically designed to evaluate the hemorrhagic risk associated with cortical brain biopsy performed during shunt surgery for iNPH.

Intracranial hemorrhage represents one of the most serious complications of shunt surgery in iNPH. The low complication rate observed in the present study compares favorably with previous reports, which have shown considerable variability in hemorrhagic complication rates [14, 17]. A notable institutional change preceding the study period was the reorganization of shunt surgery logistics in iNPH patients, with procedures scheduled during daytime hours and performed by teams specifically prepared for iNPH surgery. As the surgical technique itself remained unchanged, this organizational factor is likely a major contributor to the improved safety profile observed in the present cohort, compared with our previously reported results [6].

The diagnostic aspect of brain biopsy involves neuropathological assessment of amyloid-β and tau deposition in the neuropil. As the diagnostic utility of cortical brain biopsy in iNPH has been demonstrated in multiple previous studies [1, 7–10, 12, 15, 19, 23, 25], this aspect was beyond the scope of the present investigation. Instead, our study focused on the procedural safety of cortical brain biopsy, which has not previously been systematically evaluated.

A frequent radiological finding was the presence of minor blood within the occipital horn of the lateral ventricles, observed in 2.6% of patients. These findings were not associated with blood along the ventricular catheter tract, suggesting a mechanism distinct from direct surgical trauma. We propose that small amounts of subarachnoid blood may be transported retrogradely into the ventricular system, consistent with previously described alterations in cerebrospinal fluid dynamics in iNPH [3, 4]. Retrograde CSF flow from the subarachnoid space into the ventricles has been demonstrated using intrathecal tracer studies and may explain this characteristic distribution of intraventricular blood [22].

Traditionally, the presence of amyloid and tau deposition in the cerebral cortex, detected by brain biopsy or inferred from CSF or blood biomarkers, has been interpreted as evidence of comorbid Alzheimer’s disease in patients with iNPH [10, 16, 19]. On this basis, some authors have proposed a distinction between “true iNPH” and “neurodegenerative iNPH” [16]. However, amyloid and tau aggregation may not solely reflect coincident Alzheimer pathology but could also represent an integral component of iNPH disease progression itself. In this context, the glymphatic hypothesis provides a conceptual framework linking impaired CSF dynamics to abnormal protein clearance and subsequent cortical protein aggregation [11, 18]. This perspective raises the possibility that iNPH is not only a disorder of CSF circulation but also a disease incorporating intrinsic proteinopathy [2].

The role of cortical brain biopsy should be considered in the context of alternative biomarker strategies. For example, amyloid-PET may provide useful information in iNPH patients [10, 23, 24], but this modality has limitations related to availability, cost, and diagnostic concordance. Moreover, the role of CSF or blood biomarkers remains unclear in iNPH diagnostics [7, 20, 25, 26]. Therefore, for now, a cortical brain biopsy can be obtained during routine shunt surgery without additional procedures and provides direct histopathological confirmation, although sampling is necessarily limited to a small cortical region.

Limitations

Several limitations should be acknowledged. First, the low event rate limits statistical power to detect small differences in rare complications. Second, allocation was non-randomized and influenced by surgical team availability, which may introduce bias related to surgeon experience. Third, biopsies were performed by an experienced micro-neurosurgeon, which may limit the generalizability of the safety findings to centers with less surgical expertise. Nevertheless, surgical expertise is a relevant factor for shunt surgery itself, and our findings support that the biopsy procedure can be performed safely when appropriate expertise is available. These factors should be considered when interpreting the findings and their generalizability.

Taken together, these findings may support consideration of cortical brain biopsy in routine shunt procedures in centers with appropriate surgical expertise, particularly in the context of biomarker-driven patient stratification.

Conclusions

In this large prospective cohort, intracranial hemorrhagic complications following shunt surgery for idiopathic normal pressure hydrocephalus were rare. The addition of cortical brain biopsy was not associated with an observable increase in symptomatic or asymptomatic intracranial hemorrhage when performed by an experienced surgical team. However, given the low event rate, small but clinically relevant differences in hemorrhagic risk cannot be excluded.

Abbreviations

AD

Alzheimer’s disease

CSF

Cerebrospinal fluid

CT

Computed tomography

iNPH

Idiopathic normal pressure hydrocephalus

PET

Positron emission tomography

VA

Ventriculoatrial

VP

Ventriculoperitoneal

Author contributions

P.K.E. and A.L. contributed to the conceptualization of the study. P.K.E. and A.L. contributed to the acquisition of data. P.K.E. and A.L. contributed to the analysis and interpretation of data. P.K.E. and A.L. contributed to statistical analysis. P.K.E contributed to drafting the original text. P.K.E. and A.L. contributed to editing the text. Both authors approved the final manuscript. Correspondence and material requests: PKE.

Funding

Open access funding provided by University of Oslo (incl Oslo University Hospital) This work was supported by the foundation Stiftelsen Kristian Gerhard Jebsen through its program for translational medical research.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

This study was approved as a hospital quality improvement study (approval number 2011–6692).

Ethical standards

Ethical standards in line with the Declaration of Helsinki (1975, revised in 1983) were strictly followed.

Consent for publication

Not applicable.

Competing interest

The authors declare no competing interests.