A 7-Point Risk Stratification Tool for EVD Placement in Acute Intracerebral Hemorrhage: A Multivariable Analysis and the Development of Novel Predictive Score (EVD-ICH Score)

Abstract

Background

Intracerebral hemorrhage (ICH) is associated with high morbidity and mortality. Intraventricular extension (IVE) and hydrocephalus (HCP) frequently prompt external ventricular drain (EVD) insertion, but objective criteria to guide EVD use are lacking.

Aims

This article aims to identify clinical and radiological predictors of EVD insertion in spontaneous ICH and to develop a simple bedside scoring system (EVD-ICH score) to support decision-making.

Materials and Methods

This is a prospective observational study of 100 consecutive adults with spontaneous nontraumatic ICH admitted to a tertiary-care center (March 2023 to February 2024). Clinical and CT variables were recorded. Multivariable logistic regression identified independent predictors of EVD insertion. A points-based score was created from adjusted odds ratios and internally validated using receiver operating characteristic (ROC) analysis.

Results

Of 100 patients, predictors independently associated with EVD insertion were IVE, HCP, Glasgow coma scale (GCS) ≤ 8, hematoma volume ≥30 mL, and history of hypertension (HTN). The 7-point EVD-ICH score (IVE 2 pts, HCP 2 pts, GCS ≤8 1 pt, ICH ≥30 mL 1 pt, HTN 1 pt) achieved an area under the ROC curve (AUC) of 0.85 (95% CI: 0.78–0.92). Optimal cut-off ≥3 yielded a sensitivity of 80% and a specificity of 77%. Predicted EVD probability ranged from 8% (score 0) to 92% (score 7).

Conclusion

The EVD-ICH score provides a concise bedside tool to stratify risk of requiring EVD in spontaneous ICH. External multicenter validation and assessment of impact on patient-centered outcomes are recommended before routine adoption.

Keywords

intraventricular hemorrhage - hydrocephalus - Glasgow coma scale - scoring system - risk assessment

Introduction

Intracerebral hemorrhage (ICH) represents a severe subtype of stroke, responsible for high rates of mortality and disability worldwide.[1] Among these patients, intraventricular hemorrhage (IVH) occurs in nearly 40%[2] of cases and is consistently linked to worse prognoses, including elevated intracranial pressure (ICP), hydrocephalus (HCP), and early neurological deterioration.[3] External ventricular drains (EVDs) are frequently used in patients with IVH[4] to alleviate elevated ICP, enable cerebrospinal fluid (CSF) diversion, and, in some instances, facilitate the clearance of intraventricular blood. While EVD placement has been associated with reduced short-term mortality in selected populations, its influence on functional recovery remains inconclusive.[5] Moreover, risks such as infection and catheter-related hemorrhage are not negligible[6] and highlight the need for careful patient selection.

Despite the potential life-saving benefits of EVDs, their use remains inconsistent.[6] The absence of standardized criteria[7] [8] [9] or clinical guidelines has led to considerable variability in practice across centers and among surgeons. This variability not only complicates decision-making but also introduces potential bias in clinical trials involving patients with ICH and IVH, where EVD placement is often left to the treating team's discretion.

Given the heterogeneity of clinical presentations in ICH,[10] an individualized, evidence-based approach to determining the need for EVD placement is essential. To address this gap, we conducted a prospective analysis at a large academic institution with the following goals: to identify clinical, demographic, and radiographic predictors that influence EVD utilization; to evaluate the association between EVD placement and early clinical outcomes; and, most importantly, to develop a novel, point-based risk stratification tool to guide standardized and data-driven decision-making for EVD use in acute ICH. This study represents one of the most comprehensive single-center investigations of EVD practices in ICH patients with and without IVH.

Materials and Methods

This prospective observational study was conducted at Sher-i-Kashmir Institute of Medical Sciences, a major tertiary-care center located in northern India. The study spanned a 12-month period from March 2023 to February 2024 and aimed to predict clinical predictors related to EVD insertion in patients presenting with spontaneous ICH.

Patient Selection

A total of 100 adult patients (≥18 years) with nontraumatic spontaneous ICH were included consecutively following presentation to the emergency department. Inclusion criteria included the presence of ICH confirmed on noncontrast computed tomography (NCCT) of the brain and the absence of any recent head trauma. Patients were excluded if they had posttraumatic hematomas, intracranial tumors with hemorrhage, hemorrhagic transformation of ischemic stroke, anticoagulant- or thrombolytic-associated bleeds, vascular anomalies such as arteriovenous malformations or aneurysms, or venous sinus thrombosis.

Data Collection

Upon admission, all patients underwent a standardized evaluation protocol. Initial clinical assessment included detailed demographic profiling (age, sex, residence), medical history (hypertension, diabetes, ischemic heart disease, substance abuse, prior strokes), and current medication use, especially antihypertensives, antiplatelets, and anticoagulants. Vital signs, including blood pressure at presentation, were recorded. The Glasgow coma scale (GCS) was used to quantify neurological status.[11]

Blood samples were collected for laboratory. The decision to place an EVD was made by the attending neurosurgical team based on clinical and radiographic criteria, including decreased level of consciousness (GCS ≤ 8), obstructive HCP, significant IVH burden, or need for direct ICP monitoring. EVDs were inserted at Kocher's point[12] as shown in [Fig. 1] under sterile technique into the lateral ventricle (right-sided by default unless anatomically contraindicated). CSF samples were periodically obtained and analyzed to monitor for infection.

Fig. 1 Demonstrating external ventricular drain insertion at Kocher's point located ∼2.5–3 cm lateral to the midline and 1 cm anterior to the coronal suture. From Kocher's point, the catheter is directed medially and slightly caudally so its tip follows a straight trajectory into the ipsilateral frontal horn and toward the region of the foramen of Monro. We orient the catheter toward a target point roughly in line with the ipsilateral medial canthus or slightly toward the ipsilateral tragus, advancing gently until free cerebrospinal fluid is obtained or the desired depth is reached.

Weaning protocols[13] involved gradual elevation of the EVD system to 15 cm above the external auditory meatus, followed by a 24-hour clamp trial. If patients tolerated the clamp without neurological deterioration or elevated ICP, a confirmatory CT scan was obtained, after which the EVD was removed. Patients unable to tolerate EVD weaning were evaluated for permanent CSF diversion via ventriculoperitoneal shunt placement.[14]

Results

The study compared demographic, clinical, and radiological characteristics between patients who underwent EVD insertion (EVD group) and those managed conservatively (non-EVD group). The EVD group had a mean age of 62.4 ± 12.1 years compared with 65.8 ± 14.3 years in the non-EVD group (p = 0.18), indicating no significant age difference between groups. Sex distribution was similar, with 58% males in the EVD group versus 52% in the non-EVD group (p = 0.45; [Table 1]).

Table 1 Baseline clinical and radiological characteristics of EVD and non-EVD patient groups

Variable	EVD group (n = 41)	Non-EVD group (n = 59)	p-Value
Age (mean ± SD)	62.4 ± 12.1	65.8 ± 14.3	0.18
Sex (% male)	58%	52%	0.45
Hypertension (%)	78%	65%	0.03
GCS (median [IQR])	9[7] [8] [9] [10] [11] [15]	12[10] [11] [15] [16] [17] [18]	<0.001
ICH volume (mean ± SD)	32.5 ± 18.2	18.7 ± 12.4	<0.001
Smoker (%)	45%	38%	0.32
Laterality (left %)	55%	48%	0.25
Basal ganglia (%)	62%	40%	0.01
Thalamus (%)	25%	15%	0.08
Lobar (%)	18%	12%	0.22
IVH (%)	68%	32%	<0.001
HCP (%)	40%	15%	0.002
ICH score (median [IQR])	2[1] [2] [3]	1 [0–2]	<0.001

Abbreviations: EVD, external ventricular drain; GCS, Glasgow coma scale; HCP, hydrocephalus; ICH, intracerebral hemorrhage; IQR, interquartile range; IVH, intraventricular hemorrhage; SD, standard deviation.

Several clinically significant differences emerged between the groups. Hypertension was more prevalent in the EVD group (78 vs. 65%, p = 0.03), suggesting its potential role in disease severity. Neurological status, as measured by GCS, was significantly worse in the EVD group (median: 9; interquartile range [IQR]: 7–12]) compared with the non-EVD group (median: 12 [IQR: 10–15], p < 0.001). Hematoma volumes were substantially larger in the EVD group (32.5 ± 18.2 mL vs 18.7 ± 12.4 mL, p < 0.001), reinforcing the association between hemorrhage size and need for intervention.

Radiological findings showed important distinctions. Intraventricular extension (IVE) was markedly more common in the EVD group (68 vs. 32%, p < 0.001), as was HCP (40 vs. 15%, p = 0.002). Location analysis revealed basal ganglia involvement in 62% of EVD cases versus 40% of non-EVD cases (p = 0.01), while thalamic (25 vs. 15%, p = 0.08) and lobar (18 vs. 12%, p = 0.22) hemorrhages showed nonsignificant trends. The ICH score, a composite severity measure, was significantly higher in the EVD group (median: 2 investigations, including complete blood count, renal and liver function tests, serum electrolytes, coagulation profile, arterial blood gases, and lipid panel. Imaging with NCCT was performed to assess ICH location,[15] volume (using the ABC/2 method[16]), midline shift, presence of IVE, and HCP. The ICH score was calculated for all patients[17] to aid prognostic assessment. Patients were managed according to institutional protocols, which included blood pressure control, ICP management with mannitol,[18] seizure prophylaxis if indicated, fluid and electrolyte balance, and general supportive care, including airway protection, nutrition via nasogastric feeding, and pressure sore prevention. EVD placement and monitoring [IQR: 1–3]) were compared with the non-EVD group (median: 1 [IQR: 0–2], p < 0.001).

Other variables, including smoking status (45 vs. 38%, p = 0.32) and hemorrhage laterality (55% left-sided in EVD group vs. 48%, p = 0.25), did not differ significantly between groups. These findings collectively demonstrate that EVD placement was primarily driven by factors indicating greater disease severity: lower GCS, larger hematoma volume, IVE, HCP, and higher ICH scores, while demographic factors showed minimal influence on treatment decisions.

The multivariate logistic regression analysis identified five key predictors significantly associated with EVD insertion in patients with ICH. IVE emerged as the strongest independent predictor, with an adjusted odds ratio (aOR) of 4.92 (95% CI: 1.65–14.67, p = 0.004), indicating patients with IVE had nearly five times greater odds of requiring EVD placement compared with those without ventricular involvement. HCP also demonstrated strong predictive value, showing a fourfold increased likelihood of EVD need (aOR = 4.15, 95% CI: 1.22–14.11, p = 0.023; [Table 2]).

Table 2 Top-five independent predictors of EVD insertion

Predictor variable	Adjusted odds ratio (aOR)	95% CI for aOR	p-Value
IVE	4.92	1.65–14.67	0.004
HCP	4.15	1.22–14.11	0.023
GCS (per 1-point decrease)	0.80	0.68–0.94	0.006
ICH volume (per 1 mL)	1.05	1.01–1.09	0.012
Hypertension	2.85	0.91–8.93	0.072

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; EVD, external ventricular drain; GCS, Glasgow coma scale; HCP, hydrocephalus; ICH, intracerebral hemorrhage; IVE, intraventricular extension.

Neurological status, as measured by the GCS, showed a significant inverse relationship with EVD requirement (aOR = 0.80 per 1-point decrease, 95% CI: 0.68–0.94, p = 0.006), meaning each 1-point reduction in GCS score was associated with a 25% increase in the odds of EVD insertion. Hematoma volume exhibited a dose–response relationship, with each 1-mL increase in ICH volume being associated with a 5% higher odds of EVD placement (aOR = 1.05, 95% CI: 1.01–1.09, p = 0.012). Hypertension showed a clinically relevant though statistically borderline association (aOR = 2.85, 95% CI: 0.91–8.93, p = 0.072), suggesting hypertensive patients may have nearly three times greater odds of requiring EVD, though this finding warrants further investigation in larger studies.

The model demonstrated excellent discrimination (AUC = 0.84) and goodness-of-fit (Hosmer-Lemeshow, p = 0.67), explaining ∼50% of the variance in EVD decision-making (Nagelkerke R ² = 0.50). These results collectively indicate that the presence of intraventricular blood, HCP, depressed consciousness, and larger hematoma volumes are the most robust clinical indicators for EVD placement in ICH patients.

Scoring System Based on Predictors (EVD-ICH Score)

We assign points to each predictor based on its aOR from the multivariate analysis.

The study developed a practical 7-point scoring system to predict EVD insertion need based on multivariate logistic regression results. The strongest predictors, IVE and HCP, were each assigned 2 points due to their substantial aORs (4.92 and 4.15, respectively), reflecting their critical role in clinical decision-making. Severe neurological impairment (GCS ≤8) and large hematoma volume (≥30 mL) received 1 point each, corresponding to their significant but relatively weaker predictive values (aOR = 0.80 and 1.05). Hypertension was also allocated 1 point, acknowledging its borderline statistical significance (aOR = 2.85, p = 0.072) but clinical relevance in hematoma expansion risk ([Table 3]).

Table 3 Development of the EVD-ICH score: points assigned to independent predictors based on multivariate odds ratios

Predictor	aOR	Points assigned
IVE	4.92	2
HCP	4.15	2
GCS ≤8	0.80	1
ICH volume ≥30 mL	1.05	1
HTN	2.85	1

Abbreviations: aOR, adjusted odds ratio; EVD, external ventricular drain; GCS, Glasgow coma scale; HCP, hydrocephalus; HTN, hypertension; ICH, intracerebral hemorrhage; IVE, intraventricular extension.

Note: Total possible score = 7.

This weighted scoring approach translates complex statistical relationships into an actionable clinical tool, where higher total scores indicate progressively greater likelihood of requiring EVD intervention. The maximum possible score of 7 points incorporates all major risk factors identified in the analysis, with IVE and HCP contributing disproportionately due to their dominant predictive power. The system's design emphasizes both statistical validity and practical utility at the bedside, enabling rapid risk stratification while accounting for the relative importance of each predictor variable. The clear point thresholds (0–7) facilitate straightforward clinical implementation while maintaining the nuanced relationships between predictors evident in the original regression model.

The predictive performance of the EVD scoring system was rigorously evaluated through comprehensive validation metrics. Receiver operating characteristic (ROC) analysis demonstrated excellent discriminative ability, with an area under the curve (AUC) of 0.85 (95% CI: 0.78–0.92), indicating strong capability to differentiate between patients requiring versus not requiring EVD insertion ([Fig. 2]). The optimal cutoff was established at ≥3 points using Youden's index, achieving balanced sensitivity (80%) and specificity (77%) for clinical decision-making. The model showed a high positive predictive value (82%) for identifying EVD candidates and a reliable negative predictive value (75%) for ruling out the need for drainage.

Fig. 2 Receiver operating characteristic curve.

Goodness-of-fit assessments confirmed the model's robust calibration with clinical outcomes. The nonsignificant Hosmer-Lemeshow test result (χ ² = 4.32, p = 0.83) indicated excellent agreement between predicted probabilities and observed event rates. The Nagelkerke R ² of 0.52 revealed that the scoring system explains 52% of the variance in EVD decision-making, while the Brier score of 0.14 further supported good model calibration. These validation metrics collectively demonstrate that the scoring system maintains strong predictive accuracy while remaining well-calibrated to actual clinical outcomes, though the explained variance suggests potential for further refinement through additional predictor variables in future iterations.

The study established a clinically actionable risk stratification system for EVD insertion based on the 7-point scoring model. Patients scoring 0–1 points (very low risk) demonstrated less than 15% probability of requiring EVD, warranting conservative management with serial neurological assessments and routine follow-up imaging. The low risk category (2 points, 28% probability) suggested closer monitoring, including consideration of ICP monitoring for patients with GCS ≤12 and more frequent CT surveillance (12-hour intervals).

For moderate risk patients (3–4 points, 45–62% probability), the protocol recommended active preparation for potential EVD placement and heightened vigilance for neurological deterioration, prompting emergency reimaging if clinical decline occurred. High-risk cases (5–6 points, 76–86% probability) necessitated urgent EVD placement, ICU-level care, and comprehensive preparation for possible surgical evacuation, including maintenance of adequate cerebral perfusion pressure.

The critical-risk group (7 points, 92% probability) mandated emergent neurosurgical intervention with simultaneous EVD placement and OR team activation, coupled with aggressive ICP control measures ([Tables 4] and [5]).

Table 4 Risk stratification for EVD insertion

Total score	Probability of EVD need	Risk category
0	8%	Very low
1	15%	Low
2	28%	Moderate-low
3	45%	Moderate
4	62%	Moderate-high
5	76%	High
6	86%	Very high
7	92%	Critical

Abbreviation: EVD, external ventricular drain.

Table 5 EVD-ICH risk-based clinical framework

Score	Risk category	EVD probability
0–1	Very low risk	<15%
2	Low risk	28%
3–4	Moderate risk	45–62%
5–6	High risk	76–86%
7	Critical risk	92%

Abbreviations: EVD, external ventricular drain; ICH, intracerebral hemorrhage.

This tiered approach provides clear clinical guidance while maintaining flexibility for individualized patient assessment, with intervention intensity escalating appropriately with increasing risk scores. The stratification system effectively translates statistical predictions into practical management pathways, optimizing both resource utilization and patient outcomes.

Discussion

In this study, we developed and validated a novel 7-point risk stratification tool—the EVD-ICH score—to predict the need for EVD placement in patients presenting with acute ICH. Through a robust multivariable analysis, we identified five independent predictors of EVD insertion: IVE, HCP, low GCS, large ICH volume, and hypertension. These factors were translated into a weighted, bedside-friendly scoring system with excellent discriminative ability (AUC: 0.85), practical risk thresholds, and direct clinical utility.

Although it has been suggested that intraventricular rupture of an ICH might help reduce the mass effect exerted by a large hematoma on adjacent brain structures, studies have consistently shown that ICH extension into the ventricles independently predicts worse outcomes in affected patients.[19] [20]

In a cohort of patients with supratentorial ICH and IVH, Young et al[21] identified ventricular blood volume as a strong predictor of poor outcome, noting that patients with more than 20 cc of intraventricular blood generally had an unfavorable prognosis. In contrast, another study focusing on patients with large-volume IVH found that the underlying cause of the hemorrhage was a more significant determinant of outcome than the volume of IVH itself.[22] Tuhrim et al[19] conducted a prospective study to evaluate the prognostic and pathophysiological implications of IVH extension in ICH. They found that 30-day mortality was significantly higher in patients with IVH, with a clear association between IVH volume and poor outcome. Importantly, this association remained significant even after adjusting for HCP and the size of the accompanying ICH, thereby establishing IVH volume as an independent predictor of poor outcome.

The presence of IVH remains one of the most consistent predictors of EVD utilization, as confirmed by several studies, including those by Lovasik et al[2] and Hughes et al.[23] In our model, IVE emerged as a strong independent driver for EVD placement, aligning with findings by Lovasik et al who noted that, irrespective of other symptoms, the presence of IVH was the most decisive factor for EVD insertion. Similarly, Hughes et al emphasized that patients with panventricular IVH and higher mGraeb scores[24] were significantly more likely to require EVD.[25] Our data reinforce this association and consolidate IVE as a key input in our scoring system.

HCP, particularly in the setting of IVH, was also found to significantly correlate with EVD need. This echoes the findings of Warren et al[5] and Hughes et al, who described a high likelihood of EVD requirement in patients with obstructive or panventricular HCP. However, as Hughes et al noted, the development of HCP is variable and not always predictable based on the extent or location of IVH alone, emphasizing the need for real-time clinical assessment and standardized triage tools such as the EVD-ICH score.

Low GCS (≤8) has repeatedly surfaced as a robust indicator for EVD consideration.[26] In our study, as in the studies by Warren et al, Lovasik et al, and Mohamed et al,[27] diminished consciousness strongly predicted EVD use. Interestingly, although some may perceive extremely low GCS scores as suggestive of poor prognosis and therefore potential candidates for conservative management, our findings—as well as those of Warren et al—suggest that clinicians often proceed with EVD placement even in such patients, likely to prevent further deterioration or to preserve salvageable neurologic function.

ICH volume ≥30 mL was another significant predictor in our cohort. While previous literature[5] [28] suggests that patients with large IVH volumes and smaller ICH volumes are more likely to receive EVDs, our findings support the notion that a critical ICH volume threshold—especially when associated with intraventricular involvement or mass effect—triggers consideration for CSF diversion. Additionally, large volumes may be more likely to cause secondary HCP or midline shift, both indications for EVD.

Hypertension, while not often emphasized in prior EVD stratification studies, emerged in our analysis as an independent contributor to EVD likelihood. This may be due to the role of elevated blood pressure in patients with IVH contributing to both hematoma expansion and IVH propagation.[5] [29]

Our model is also timely, given the absence of definitive guidelines for EVD placement in spontaneous ICH. As reflected in the STICH[30] and NovoSeven trials,[31] EVDs were underutilized—placed in only ∼10% of patients with IVH—potentially due to the lack of consensus or predictive tools. Our score offers a standardized, objective measure to support early neurosurgical decisions, especially in settings where practice variability and junior-level residents may lead to inconsistent EVD use.[32]

However, it is important to contextualize EVD placement within its broader impact on patient outcomes. While studies have demonstrated that EVD use can reduce mortality in IVH and ICH patients, there is less evidence that it translates into improved long-term functional outcomes. Several trials have reported a paradox wherein life is prolonged, but with significant neurologic morbidity. Mohamed et al[27] even observed that conservative management may yield better outcomes in patients with relatively preserved consciousness (GCS > 10), cautioning against indiscriminate EVD use.

Conclusion

This study developed and validated the EVD-ICH score, a 7-point tool for predicting the need for EVD placement in patients with acute ICH. The score incorporates IVE, HCP, low GCS, large hematoma volume (≥30 mL), and hypertension as independent predictors, achieving strong discriminative performance (AUC: 0.85). The EVD-ICH score provides a standardized, bedside approach to guide neurosurgeons and intensivists in identifying patients who may benefit from early CSF diversion.

Conflict of Interest

None declared.

Authors' Contributions

All the authors have read and approved the manuscript.

Ethical Approval

Being an observational prospective study, the study was exempted by the Institutional Ethics Committee of SKIMS. The study was conducted in accordance with the Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects.

Patients' Consent

A written consent to participate in the study was taken from the patient or next of kin.

• References

• 1 Wang ZW, Wan MP, Tai JH, Wang Y, Yin MY. Global regional and national burden of intracerebral hemorrhage between 1990 and 2021. Sci Rep 2025; 15 (01) 3624
• 2 Lovasik BP, McCracken DJ, McCracken CE. et al. Effect of external ventricular drain use in intracerebral hemorrhage. World Neurosurg 2016; 94: 309-318
• 3 Hinson HE, Hanley DF, Ziai WC. Management of intraventricular hemorrhage. Curr Neurol Neurosci Rep 2010; 10 (02) 73-82
• 4 Dey M, Jaffe J, Stadnik A, Awad IA. External ventricular drainage for intraventricular hemorrhage. Curr Neurol Neurosci Rep 2012; 12 (01) 24-33
• 5 Warren AD, Li Q, Schwab K. et al. External ventricular drain placement is associated with lower mortality after intracerebral hemorrhage with intraventricular hemorrhage. Int J Emerg Med 2022; 15 (01) 51
• 6 Yuen J, Selbi W, Muquit S, Berei T. Complication rates of external ventricular drain insertion by surgeons of different experience. Ann R Coll Surg Engl 2018; 100 (03) 221-225
• 7 Nieuwkamp DJ, de Gans K, Rinkel GJ, Algra A. Treatment and outcome of severe intraventricular extension in patients with subarachnoid or intracerebral hemorrhage: a systematic review of the literature. J Neurol 2000; 247 (02) 117-121
• 8 Engelhard HH, Andrews CO, Slavin KV, Charbel FT. Current management of intraventricular hemorrhage. Surg Neurol 2003; 60 (01) 15-21 , discussion 21–22
• 9 Ruscalleda J, Peiró A. Prognostic factors in intraparenchymatous hematoma with ventricular hemorrhage. Neuroradiology 1986; 28 (01) 34-37
• 10 Morotti A, Goldstein JN. Diagnosis and management of acute intracerebral hemorrhage. Emerg Med Clin North Am 2016; 34 (04) 883-899
• 11 Teasdale G, Maas A, Lecky F, Manley G, Stocchetti N, Murray G. The Glasgow Coma Scale at 40 years: standing the test of time. Lancet Neurol 2014; 13 (08) 844-854
• 12 Munakomi S, Das JM. Ventriculostomy. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025
• 13 Kothari SA, Siddiq MS, Rahimi S, Vale F, Shah M, Garcia KA. Standardized criteria to initiate external ventricular drain (EVD) weaning in neurological intensive care to increase safety and reduce shunt need. Cureus 2024; 16 (04) e58362
• 14 Li CR, Yen CM, Yang MY, Cheng WY, Shen CC, Liu SY. Predictive factors for shunt dependency in patients with spontaneous intraventricular hemorrhage. Sci Rep 2024; 14 (01) 26462
• 15 Delgado Almandoz JE, Schaefer PW, Goldstein JN. et al. Practical scoring system for the identification of patients with intracerebral hemorrhage at highest risk of harboring an underlying vascular etiology: the Secondary Intracerebral Hemorrhage Score. AJNR Am J Neuroradiol 2010; 31 (09) 1653-1660
• 16 Hakimi R, Garg A. Imaging of hemorrhagic stroke. Continuum (Minneap Minn) 2016; 22 (5, Neuroimaging): 1424-1450
• 17 Hemphill III JC, Bonovich DC, Besmertis L, Manley GT, Johnston SC. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke 2001; 32 (04) 891-897
• 18 Patel P, Patel R, Thorell W. Mannitol. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025
• 19 Tuhrim S, Horowitz DR, Sacher M, Godbold JH. Volume of ventricular blood is an important determinant of outcome in supratentorial intracerebral hemorrhage. Crit Care Med 1999; 27 (03) 617-621
• 20 Bhattathiri PS, Gregson B, Prasad KS, Mendelow AD. STICH Investigators. Intraventricular hemorrhage and hydrocephalus after spontaneous intracerebral hemorrhage: results from the STICH trial. Acta Neurochir Suppl (Wien) 2006; 96: 65-68
• 21 Young WB, Lee KP, Pessin MS, Kwan ES, Rand WM, Caplan LR. Prognostic significance of ventricular blood in supratentorial hemorrhage: a volumetric study. Neurology 1990; 40 (04) 616-619
• 22 Roos YB, Hasan D, Vermeulen M. Outcome in patients with large intraventricular haemorrhages: a volumetric study. J Neurol Neurosurg Psychiatry 1995; 58 (05) 622-624
• 23 Hughes JD, Puffer R, Rabinstein AA. Risk factors for hydrocephalus requiring external ventricular drainage in patients with intraventricular hemorrhage. J Neurosurg 2015; 123 (06) 1439-1446
• 24 Morgan TC. et al. The Modified Graeb Score: an enhanced tool for intraventricular hemorrhage measurement and outcome prediction. Stroke 2013; 44 (03) 635-641
• 25 Chau CYC, Mediratta S, McKie MA. et al. Optimal timing of external ventricular drainage after severe traumatic brain injury: a systematic review. J Clin Med 2020; 9 (06) E1564
• 26 Mohamed AA, Ezzat AAM, Abdul-Sh MMH. Comparative study between conservative and surgical management of intraventricular hemorrhage. Med J Cairo Univ 2022; 90 (09) 1335-1339
• 27 Mohamed D, Mostafa H, Soliman M, Meleegy M. Short-term outcomes of conservative management versus external ventricular drainage in adults with spontaneous intraventricular hemorrhage: a mixed observational cohort study. Cureus 2026; 18 (01) e101220
• 28 Herrick DB, Ullman N, Nekoovaght-Tak S. et al. Determinants of external ventricular drain placement and associated outcomes in patients with spontaneous intraventricular hemorrhage. Neurocrit Care 2014; 21 (03) 426-434
• 29 Ohwaki K, Yano E, Nagashima H, Hirata M, Nakagomi T, Tamura A. Blood pressure management in acute intracerebral hemorrhage: relationship between elevated blood pressure and hematoma enlargement. Stroke 2004; 35 (06) 1364-1367
• 30 Mendelow AD. et al. Early surgery versus initial conservative treatment in spontaneous supratentorial intracerebral haematomas: results from the STICH trial. Lancet 2005; 365 (9457) 387-397
• 31 Levi M, Levy JH, Andersen HF, Truloff D. Safety of recombinant activated factor VII in randomized clinical trials. N Engl J Med 2010; 363 (19) 1791-1800
• 32 Zakaria J, Jusue-Torres I, Frazzetta J. et al. Effectiveness of a standardized external ventricular drain placement protocol for infection control. World Neurosurg 2021; 151: e771-e777

Publication History

Article published online:
03 February 2026