Ambulatory intracranial pressure in humans: ICP increases during movement between body positions.

Introduction: Positional changes in intracranial pressure (ICP) have been described in humans when measured over minutes or hours in a static posture, with ICP higher when lying supine than when sitting or standing upright. However, humans are often ambulant with frequent changes in position self-generated by active movement. Research question: We explored how ICP changes during movement between body positions. Material and methods: Sixty-two patients undergoing clinical ICP monitoring were recruited. Patients were relatively well, ambulatory and of mixed age, body habitus and pathology. We instructed patients to move back and forth between sitting and standing or lying and sitting positions at 20 s intervals after an initial 60s at rest. We simultaneously measured body position kinematics from inertial measurement units and ICP from an intraparenchymal probe at 100 Hz. Results: ICP increased transiently during movements beyond the level expected by body position alone. The amplitude of the increase varied between participants but was on average ∼5 mmHg during sit-to-stand, stand-to-sit and sit-to-lie movements and 10.8 mmHg [95%CI: 9.3,12.4] during lie-to-sit movements. The amplitude increased slightly with age, was greater in males, and increased with median 24-h ICP. For lie-to-sit and sit-to-lie movements, higher BMI was associated with greater mid-movement increase (ß = 0.99 [0.78,1.20]; ß = 0.49 [0.34,0.64], respectively). Discussion and conclusion: ICP increases during movement between body positions. The amplitude of the increase in ICP varies with type of movement, age, sex, and BMI. This could be a marker of disturbed ICP dynamics and may be particularly relevant for patients with CSF-diverting shunts in situ.

Chemical angioplasty vs. balloon plus chemical angioplasty for delayed cerebral ischemia: a pilot study of PbtO2 outcomes.

BACKGROUND: Delayed cerebral ischaemia (DCI) is a major cause of morbidity and mortality after aneurysmal subarachnoid haemorrhage (aSAH). Chemical angioplasty (CA) and transluminal balloon angioplasty (TBA) are used to treat patients with refractory vasospasm causing DCI. Multi-modal monitoring including brain tissue oxygenation (PbtO2) is routinely used at this centre for early detection and management of DCI following aSAH. In this single-centre pilot study, we are comparing these two treatment modalities and their effects on PbtO2. METHODS: Retrospective case series of patients with DCI who had PbtO2 monitoring as part of their multimodality monitoring and underwent either CA or TBA combined with CA. PbtO2 values were recorded from intra-parenchymal Raumedic NEUROVENT-PTO® probes. Data were continuously collected and downloaded as second-by-second data. Comparisons were made between pre-angioplasty PbtO2 and post-angioplasty PbtO2 median values (4 h before angioplasty, 4 h after and 12 h after). RESULTS: There were immediate significant improvements in PbtO2 at the start of intervention in both groups. PbtO2 then increased by 13 mmHg in the CA group and 15 mmHg in the TBA plus CA group in the first 4 h post-intervention. This improvement in PbtO2 was sustained for the TBA plus CA group but not the CA group. CONCLUSION: Combined balloon plus chemical angioplasty results in more sustained improvement in brain tissue oxygenation compared with chemical angioplasty alone. Our findings suggest that PbtO2 is a useful tool for monitoring the response to angioplasty in vasospasm.

The Impact of Intracranial Pressure Telesensors: An Observational Propensity-Matched Control Analysis of Service Demand and Costs.

BACKGROUND AND OBJECTIVES: Implantable telemetric intracranial pressure (ICP) sensors (telesensors) enable routine, noninvasive ICP feedback, aiding clinical decision-making and attribution of pressure-related symptoms in patients with cerebrospinal fluid shunt systems. Here, we aim to explore the impact of these devices on service demand and costs in patients with adult hydrocephalus. METHODS: We performed an observational propensity-matched control study, comparing patients who had an MScio/Sensor Reservoir (Christoph Miethke, GmbH & Co) against those with a nontelemetric reservoir inserted between March 2016 and March 2018. Patients were matched on demographics, diagnosis, shunt-type, and revision status. Service usage was recorded with frequencies of neurosurgical admissions, outpatient clinics, scans, and further surgical procedures in the 2 years before and after shunt insertion. RESULTS: In total, 136 patients, 73 telesensors, and 63 controls were included in this study (48 matched pairs). Telesensor use led to a significant decrease in neurosurgical inpatient admissions, radiographic encounters, and procedures including ICP monitoring. After multivariate adjustment, the mean cumulative saving after 2 years was £5236 ($6338) in telesensor patients (£5498 on matched pair analysis). On break-even analysis, cost-savings were likely to be achieved within 8 months of clinical use, postimplantation. Telesensor patients also experienced a significant reduction in imaging-associated radiation (4 mSv) over 2 years. CONCLUSION: The findings of this exploratory study reveal that telesensor implantation is associated with reduced service demand and provides net financial savings from an institutional perspective. Moreover, telesensor patients required fewer appointments, invasive procedures, and had less radiation exposure, indicating an improvement in both their experience and safety.

Ventriculo-atrial shunt in idiopathic intracranial hypertension.

PURPOSE: CSF diversion  is a recognised intervention in idiopathic intracranial hypertension (IIH), particularly in the presence of vision-threatening papilledema. Although ventriculo-atrial (VA) shunt insertion is a routine neurosurgical procedure, ventriculoperitoneal and lumboperitoneal shunts have been mostly used in this particular indication. This study aims to look at a single centre's experience with VA shunts in idiopathic intracranial hypertension (IIH). METHODS: Retrospective case series with a review of electronic records over a 10-year period; exclusion criteria were duplication of same shunt insertion, no VA shunt insertion, paediatric patients and indication other than IIH. Notes were reviewed for demographics, shunt survival (defined by time prior to revision) and reasons for revision. RESULTS: Eight VA shunt procedures were identified in 6 patients (mean age at insertion 34 ± 10 years) with a mean follow-up of 58 ± 25 months. All shunts were secondary procedures; 2 revisions from lumbo-pleural, 2 from ventriculopleural, 2 from ventriculoatrial and one each from ventriculoperitoneal and combined lumbo-/ventriculoperitoneal. At 50 months, 75% of VA shunts had survived, compared to only 58.3% of VPleural shunts in patients with IIH. Revisions were required due to acute intracranial bleed (1 case)-revised at day 1, and thrombus at distal site (1 case)-revised at day 57. Both shunts were later reinserted. From the latest clinic letters, all patients had their treatment optimised with this procedure, although only two patients had documented resolved papilloedema post-procedure. CONCLUSIONS: Ventriculo-atrial shunts are a safe and efficacious alternative option for CSF diversion in IIH. In this series, only 1 shunt was revised for a VA shunt-specific complication.

Classification of Chronic Hydrocephalus in Adults: A Systematic Review and Analysis.

BACKGROUND: Chronic hydrocephalus in adults (CHiA) includes all nonacute forms of hydrocephalus occurring in adulthood. It covers a spectrum of disorders. Some of these have relatively agreed on definitions, while others are less well characterized. The existing medical classification systems lack adequate structure and are neither clinically oriented nor easy to use, which severely hampers research and clinical care efforts. METHODS: A systematic literature review and data analysis were performed, focusing on the terms "adult hydrocephalus" and "classification," using the PubMed, Scopus, and Cochrane Library databases. Data on terminology, definitions, patient demographics, symptom duration, and clinical presentations were extracted, analyzed, and compiled. A Delphi process was followed to define CHiA disorders. RESULTS: A total of 33 studies collectively used 48 terms to define various CHiA disorders. Different terms were used to describe similar conditions. CHiA disorders were found to be clustered into 7 distinctive clinical entities based on the clinical characteristics. CONCLUSIONS: An evidence-based new clinical classification for CHiA is suggested. Our review identified gaps in knowledge and areas for further research.

Can clinicians correctly predict intracranial pressure state based on clinical symptoms alone in shunted patients?

BACKGROUND: Clinicians are well-versed in the classical symptoms of low vs. high intracranial pressure (ICP). However, symptoms may not be as predictable of ICP state in shunted patients with chronic symptoms. In this study, we assess whether clinicians can predict high vs. low ICP state in chronically symptomatic shunted patients without any diagnostic clues. METHODS: A detailed retrospective analysis was performed on 259 patients undergoing ICP monitoring. A total of 17 patients who had a ventriculoperitoneal shunt were identified, with a suspected chronic abnormal ICP state based only on clinical symptoms. Patients with investigations guiding towards a likely pressure state were excluded, e.g., imaging or ophthalmological findings suggestive of ICP state. RESULTS: Clinical suspicion of ICP state was incorrect in 16 out of 17 cases (P<0.05). The symptoms described by patients were suggestive of abnormal ICP states; however, 13 out of 17 cases demonstrated ICP within the normal range (-1.3 to 5.3 mmHg). Three patients with occipital headaches worse on standing, typical of low-pressure symptoms, were in fact shown to have ICP above 10.0 mmHg. CONCLUSIONS: This study casts doubt on the utility of classic symptoms in diagnosing abnormal ICP state in chronically symptomatic shunted patients with equivocal adjuncts. Additionally, it highlights the importance of ICP monitoring for this patient group.

Theophylline, a drug efficient to increase intracranial pressure. Case report and review of literature.

Introduction: Limited information is available regarding the direct effect of drugs prescribed to elevate intracranial pressure (ICP) in contrast to the abundance of evidence in support of medication utilized to lower ICP. Theophylline is a methylxanthine derivate used in the treatment of respiratory diseases with only a few randomized control trials or case reports describing pain improvement in low CSF pressure states or spontaneous intracranial hypotension (SIH). We present the case of a recorded increase in intracranial pressure under theophylline. Case report: A 23-year-old female with a complex history of hydrocephalus and syringomyelia was experiencing debilitating symptoms due to refractory intracranial hypotension. Medical management with oral theophylline was attempted for a period of three months. Intracranial pressure measurements were obtained via the telemetric pressure sensor reservoir (Miethke®, M.scio®) incorporated in the patient's ventriculoperitoneal shunt system. Results: A significant increase in intracranial pressure was recorded at therapeutic drug levels. Conclusion: This is the first report of an increase in intracranial pressure under oral theophylline medication.

The initial intracranial pressure spike phenomenon.

BACKGROUND: Elective use of intraparenchymal intracranial pressure (ICP) monitoring is a valuable resource in the investigation of hydrocephalus and other cerebrospinal fluid disorders. Our preliminary study aims to investigate ICP changes in the immediate period following dural breach, which has not yet been reported on. METHOD: This is a prospective cohort study of patients undergoing elective ICP monitoring, recruited between March and May 2022. ICP readings were obtained at opening and then at 5-min intervals for a 30-min duration. RESULTS: Ten patients were recruited, mean age 45 years, with indications of a Chiari malformation (n = 5), idiopathic intracranial hypertension (n = 3) or other ICP-related pathology (n = 2). Patients received intermittent bolus sedation (80%) vs general anaesthesia (20%). Mean opening pressure was 22.9 mmHg [± 6.0], with statistically significant decreases present every 5 min, to a total reduction of 15.2 mmHg at 20 min (p = < 0.0001), whereafter the ICP plateaued with no further statistical change. DISCUSSION: Our results highlight an intracranial opening pressure 'spike' phenomenon. This spike was 15.2 mmHg higher than the plateau, which is reached at 20 min after insertion. Several possible causes exist which require further research in larger cohorts, including sedation and pain response. Regardless of causation, this study provides key information on the use of ICP monitoring devices, guiding interpretation and when to obtain measurements.

Cerebrospinal fluid production rate in various pathological conditions: a preliminary study.

INTRODUCTION: The cerebrospinal fluid (CSF) production rate in humans is not clearly defined but is estimated to be 18-24 ml/h (Trevisi et al Croat Med J 55(4):377-387 (24); Casey and Vries Childs Nerv Syst 5(5):332-334 (8)). A frequent clinical observation is that patients often drain higher volumes of CSF than can be explained by the assumed 'normal' CSF production rate (PRcsf). In the National Hospital for Neurology and Neurosurgery PRcsf was recorded in a variety of common neurosurgical pathologies using LiquoGuard7, an automated peristaltic pump that accurately controls CSF drainage and maintains a pre-set CSF pressure. METHODS: A prospective observational study was performed from September 2021 onwards, on all patients in the National Hospital for Neurology and Neurosurgery who required CSF drainage as part of their ongoing treatment. The external drain was connected to a LiquoGuard7 pump (Möller Medical GmbH, Fulda, Germany), and the internal software of LiquoGuard7 was used to measure PRcsf. Statistical analysis used SPSS (version 25.0, IBM) by paired t test, comparing measured rates to hypothetical 'normal' CSF production rates calculated and published by Ekstedt (16-34ml/h) (Ekstedt J Neurol Neurosurg Psychiatry 41(4):345-353 (14)), assuming a similar distribution. RESULTS: PRcsf was calculated in 164 patients. Suspected normal pressure hydrocephalus (n=41): PRcsf of 79ml/h±20SD (p<0.0001). Post-surgical CSF leak (n=26): PRcsf of 90ml/h±20SD (p<0.0001). Subarachnoid haemorrhage (n=34): PRcsf of 143ml/h±9SD (p<0.0001). Intracerebral haemorrhage (n=22): PRcsf of 137ml/h±20SD (p<0.0001). Spinal lesions (n=7): PRcsf of 130ml/h±20SD (p<0.0032). Pituitary adenomas (n=10): PRcsf of 29 ml/h±9SD (p<0.049). Idiopathic intracranial hypertension (n=15): PRcsf of 86ml/h±10SD (p<0.0001). Decompensated long-standing overt ventriculomegaly (n=4): PRcsf of 65ml/h±10SD (p<0.0001). Cerebral infection (n=5): PRcsf of 90ml/h±20SD (p<0.0001). CONCLUSION: Net CSF production rate may be higher than expected in many conditions, as measured with new device LiquoGuard7 through the study of net flow rate, which may have implications for clinical decisions on CSF diversion. The conventional understanding of CSF production and circulation does not explain the findings of this study. More extensive studies are needed to validate this technique.

Predicting vasospasm risk using first presentation aneurysmal subarachnoid hemorrhage volume: A semi-automated CT image segmentation analysis using ITK-SNAP.

PURPOSE: Cerebral vasospasm following aneurysmal subarachnoid hemorrhage (aSAH) is a significant complication associated with poor neurological outcomes. We present a novel, semi-automated pipeline, implemented in the open-source medical imaging analysis software ITK-SNAP, to segment subarachnoid blood volume from initial CT head (CTH) scans and use this to predict future radiological vasospasm. METHODS: 42 patients were admitted between February 2020 and December 2021 to our tertiary neurosciences center, and whose initial referral CTH scan was used for this retrospective cohort study. Blood load was segmented using a semi-automated random forest classifier and active contour evolution implemented in ITK-SNAP. Clinical data were extracted from electronic healthcare records in order to fit models aimed at predicting radiological vasospasm risk. RESULTS: Semi-automated segmentations demonstrated excellent agreement with manual, expert-derived volumes (mean Dice coefficient = 0.92). Total normalized blood volume, extracted from CTH images at first presentation, was significantly associated with greater odds of later radiological vasospasm, increasing by approximately 7% for each additional cm3 of blood (OR = 1.069, 95% CI: 1.021-1.120; p < .005). Greater blood volume was also significantly associated with vasospasm of a higher Lindegaard ratio, of longer duration, and a greater number of discrete episodes. Total blood volume predicted radiological vasospasm with a greater accuracy as compared to the modified Fisher scale (AUC = 0.86 vs 0.70), and was of independent predictive value. CONCLUSION: Semi-automated methods provide a plausible pipeline for the segmentation of blood from CT head images in aSAH, and total blood volume is a robust, extendable predictor of radiological vasospasm, outperforming the modified Fisher scale. Greater subarachnoid blood volume significantly increases the odds of subsequent vasospasm, its time course and its severity.

ICP during head movement: significance of the venous system.

BACKGROUND: It is thought that the internal jugular veins (IJV) are the primary route for cranial venous outflow in supine position and the vertebral venous plexus when upright. Previous studies have noted a greater increase in intracranial pressure (ICP) when subjects turn their head in one direction compared to the other, but no clear cause had been investigated. We hypothesised that in the supine position, head turning and consequently obstructing the IJV draining the dominant transverse sinus (TVS) would lead to a greater rise in ICP compared to turning to the non-dominant side. METHODS: A prospective study in a large-volume neurosurgical centre. Patients undergoing continuous ICP monitoring as part of their standard clinical management were recruited. Immediate ICP was measured in different head positions (neutral, rotated to the right and left) when supine, seated, and standing. TVS dominance was established by consultant radiologist report on venous imaging. RESULTS: Twenty patients were included in the study, with a median age of 44 years. Venous system measurements revealed 85% right-sided vs 15% left-sided dominance. Immediate ICP rose more when head turning from neutral to the dominant TVS (21.93mmHg ± 4.39) vs non-dominant side (16.66mmHg ± 2.71) (p= <0.0001). There was no significant relationship in the sitting (6.08mmHg ± 3.86 vs 4.79mmHg ± 3.81, p = 0.13) or standing positions (8.74mmHg ± 4.30 vs 6.76mmHg ± 4.14, p =0.07). CONCLUSION: This study has provided further evidence that the transverse venous sinus to internal jugular system pathway is the likely primary venous drainage when supine; and quantified its effect when head turning on ICP. It may guide patient-specific nursing care and advice.

Multidisciplinary consensus guideline for the diagnosis and management of spontaneous intracranial hypotension.

BACKGROUND: We aimed to create a multidisciplinary consensus clinical guideline for best practice in the diagnosis, investigation and management of spontaneous intracranial hypotension (SIH) due to cerebrospinal fluid leak based on current evidence and consensus from a multidisciplinary specialist interest group (SIG). METHODS: A 29-member SIG was established, with members from neurology, neuroradiology, anaesthetics, neurosurgery and patient representatives. The scope and purpose of the guideline were agreed by the SIG by consensus. The SIG then developed guideline statements for a series of question topics using a modified Delphi process. This process was supported by a systematic literature review, surveys of patients and healthcare professionals and review by several international experts on SIH. RESULTS: SIH and its differential diagnoses should be considered in any patient presenting with orthostatic headache. First-line imaging should be MRI of the brain with contrast and the whole spine. First-line treatment is non-targeted epidural blood patch (EBP), which should be performed as early as possible. We provide criteria for performing myelography depending on the spine MRI result and response to EBP, and we outline principles of treatments. Recommendations for conservative management, symptomatic treatment of headache and management of complications of SIH are also provided. CONCLUSIONS: This multidisciplinary consensus clinical guideline has the potential to increase awareness of SIH among healthcare professionals, produce greater consistency in care, improve diagnostic accuracy, promote effective investigations and treatments and reduce disability attributable to SIH.

Aneurysm management in patients over 80 years old with good grade subarachnoid haemorrhage.

BACKGROUND: An increasing proportion of aneurysmal subarachnoid haemorrhage (aSAH) occurs in older patients, in whom there is widespread variability in treatment rates due to a different balance of risks. Our aim was to compare outcomes of patients over 80 years old with good grade aSAH who underwent treatment of their aneurysm with those who did not. METHODS: Adult patients with good grade aSAH admitted to tertiary regional neurosciences centres contributing to the UK and Ireland Subarachnoid Haemorrhage Database (UKISAH) and a cohort of consecutive patients admitted from three regional cohorts were included for analysis. Outcomes were functional outcome at discharge, three months and survival at discharge. RESULTS: In the UKISAH, patients whose aneurysm was treated were more likely to have a favourable outcome at discharge (OR 2.34, CI 1.12-4.91, p = .02), at three months (OR 2.29, CI 1.11-4.76, p = .04), and lower mortality (10% vs. 29%, OR 0.83, CI 0.72-0.94, p < .01). In the regional cohort, a similar pattern was seen, but after correction for frailty and comorbidity there was no difference in survival (HR 0.45, CI 0.12-1.68, p = .24) or favourable outcome at discharge (OR 0.83, CI 0.23-2.94, p = .77) and at three months (OR 1.03, CI 0.25-4.29, p = .99). CONCLUSIONS: Better early functional outcomes in those undergoing aneurysm treatment appear to be explained by differences in frailty and comorbidity. Therefore, treatment decisions in this patient group are finely balanced with no clear evidence overall of either benefit or harm in this cohort.

Predicting neurosurgical referral outcomes in patients with chronic subdural hematomas using machine learning algorithms - A multi-center feasibility study.

Background: Chronic subdural hematoma (CSDH) incidence and referral rates to neurosurgery are increasing. Accurate and automated evidence-based referral decision-support tools that can triage referrals are required. Our objective was to explore the feasibility of machine learning (ML) algorithms in predicting the outcome of a CSDH referral made to neurosurgery and to examine their reliability on external validation. Methods: Multicenter retrospective case series conducted from 2015 to 2020, analyzing all CSDH patient referrals at two neurosurgical centers in the United Kingdom. 10 independent predictor variables were analyzed to predict the binary outcome of either accepting (for surgical treatment) or rejecting the CSDH referral with the aim of conservative management. 5 ML algorithms were developed and externally tested to determine the most reliable model for deployment. Results: 1500 referrals in the internal cohort were analyzed, with 70% being rejected referrals. On a holdout set of 450 patients, the artificial neural network demonstrated an accuracy of 96.222% (94.444-97.778), an area under the receiver operating curve (AUC) of 0.951 (0.927-0.973) and a brier score loss of 0.037 (0.022-0.056). On a 1713 external validation patient cohort, the model demonstrated an AUC of 0.896 (0.878-0.912) and an accuracy of 92.294% (90.952-93.520). This model is publicly deployed: https://medmlanalytics.com/neural-analysis-model/. Conclusion: ML models can accurately predict referral outcomes and can potentially be used in clinical practice as CSDH referral decision making support tools. The growing demand in healthcare, combined with increasing digitization of health records raises the opportunity for ML algorithms to be used for decision making in complex clinical scenarios.

The benefits of automated CSF drainage in normal pressure hydrocephalus.

BACKGROUND: The commonly used cerebrospinal fluid (CSF) drainage system remains the manual drip-chamber drain. The LiquoGuard (Möller Medical GmbH, Germany) is an automated CSF management device with dual functionality, measuring intracranial pressure and automatic pressure- or volume-led CSF drainage. There is limited research for comparison of devices, particularly in the neurosurgical field, where it has potential to reshape care. OBJECTIVE: This study aims to compare manual drip-chamber drain versus LiquoGuard system, by assessing accuracy of drainage, associated morbidity and impact on length of stay. METHOD: Inclusion criteria consisted of suspected normal pressure hydrocephalus (NPH) patients undergoing extended lumbar drainage. Patients were divided into manual drain group versus automated group. RESULTS: Data was analysed from 42 patients: 31 in the manual group versus 11 in the LiquoGuard group. Volumetric over-drainage was seen in 90.3% (n = 28) versus 0% (p < 0.05), and under-drainage in 38.7% (n = 12) versus 0% (p < 0.05), in the manual and automatic group, respectively. Symptoms of over-drainage were noted in 54.8% (n = 17) of the manual group, all of which had episodes of volumetric over-drainage, versus 18.2% (n = 2) in automated group, of which neither had actual over-drainage (p < 0.05). Higher over-drainage symptoms of manual drain is likely due to increased fluctuation of CSF drainage, instead of smooth CSF drainage seen with LiquoGuard system. An increased length of stay was seen in 38.7% (n = 12) versus 9% (n = 1) (p < 0.05) in the manual and LiquoGuard group, respectively. CONCLUSION: The LiquoGuard device is a more superior way of CSF drainage in suspected NPH patients, with reduced morbidity and length of stay.

A case of recurrent flight-induced cerebrospinal fluid shunt overdrainage.

Shunted patients often complain of headaches after flights. The effect of air travel on shunt systems is unknown. We describe the case of a patient with longstanding hydrocephalus, who suffered flight-induced clinical deterioration and shunt overdrainage in two independent occasions. The patient, clinically stable for 1.5 and 5 years before each episode, reported severe headaches starting during the descent stages of the air travel. On both occasions, brain MRI imaging demonstrated pronounced ventricular size reduction. This case suggests that flight-induced shunt overdrainage can occur and should be suspected in patients with prolonged headaches and/or clinical deterioration triggered by air travel.

A survey of the radiological follow-up of unruptured intracranial aneurysms in the United Kingdom.

OBJECTIVE: Unruptured intracranial aneurysms (UIA) are common. For many the treatment risks outweigh their risk of subarachnoid haemorrhage and patients undergo surveillance imaging. There is little data to inform if and how to monitor UIAs resulting in widely varying practices. This study aimed to determine the current practice of unruptured UIA surveillance in the United Kingdom. METHODS: A questionnaire was designed to address the themes of surveillance protocols for UIA including when surveillance is initiated, how frequently it is performed, and when it is terminated. Additionally, how aneurysm growth is managed and how clinically meaningful growth is defined were explored. The questionnaire was distributed to members of the British Neurovascular Group using probability-based cluster and non-probability purposive sampling methods. RESULTS: Responses were received from 30 of the 30 (100.0%) adult neurosurgical units in the United Kingdom of which 27 (90.0%) routinely perform surveillance for aneurysm growth. Only four units had a unit policy. The mean patient age up to which a unit would initiate follow-up of a low-risk UIA was 65.4 ± 9.0 years. The time points at which imaging is performed varied widely. There was an even split between whether units use a fixed duration of follow-up or an age threshold for terminating surveillance. Forty percent of units will follow-up patients more than 5 years from diagnosis. The magnitude in the change in size that was felt to constitute growth ranged from 1 to 3mm. No units routinely used vessel wall imaging although 27 had access to 3T MRI capable of performing it. CONCLUSIONS: There is marked heterogeneity in surveillance practices between units in the United Kingdom. This study will help units better understand their practice relative to their peers and provide a framework forplanning further research on aneurysm growth.

Characterization of patients with idiopathic normal pressure hydrocephalus using natural language processing within an electronic healthcare record system.

OBJECTIVE: Idiopathic normal pressure hydrocephalus (iNPH) is an underdiagnosed, progressive, and disabling condition. Early treatment is associated with better outcomes and improved quality of life. In this paper, the authors aimed to identify features associated with patients with iNPH using natural language processing (NLP) to characterize this cohort, with the intention to later target the development of artificial intelligence-driven tools for early detection. METHODS: The electronic health records of patients with shunt-responsive iNPH were retrospectively reviewed using an NLP algorithm. Participants were selected from a prospectively maintained single-center database of patients undergoing CSF diversion for probable iNPH (March 2008-July 2020). Analysis was conducted on preoperative health records including clinic letters, referrals, and radiology reports accessed through CogStack. Clinical features were extracted from these records as SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms) concepts using a named entity recognition machine learning model. In the first phase, a base model was generated using unsupervised training on 1 million electronic health records and supervised training with 500 double-annotated documents. The model was fine-tuned to improve accuracy using 300 records from patients with iNPH double annotated by two blinded assessors. Thematic analysis of the concepts identified by the machine learning algorithm was performed, and the frequency and timing of terms were analyzed to describe this patient group. RESULTS: In total, 293 eligible patients responsive to CSF diversion were identified. The median age at CSF diversion was 75 years, with a male predominance (69% male). The algorithm performed with a high degree of precision and recall (F1 score 0.92). Thematic analysis revealed the most frequently documented symptoms related to mobility, cognitive impairment, and falls or balance. The most frequent comorbidities were related to cardiovascular and hematological problems. CONCLUSIONS: This model demonstrates accurate, automated recognition of iNPH features from medical records. Opportunities for translation include detecting patients with undiagnosed iNPH from primary care records, with the aim to ultimately improve outcomes for these patients through artificial intelligence-driven early detection of iNPH and prompt treatment.