Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size

Clio Rubinos; Soon Bin Kwon; Murad Megjhani; Kalijah Terilli; Brenda Wong; Lizbeth Cespedes; Jenna Ford; Renz Reyes; Hannah Kirsch; Ayham Alkhachroum; Angela Velazquez; David Roh; Sachin Agarwal; Jan Claassen; E. Sander Connolly; Soojin Park

doi:10.1007/s12028-022-01538-8

Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size

Clio Rubinos, Soon Bin Kwon, Murad Megjhani, Kalijah Terilli, Brenda Wong, Lizbeth Cespedes, Jenna Ford, Renz Reyes, Hannah Kirsch, Ayham Alkhachroum, Angela Velazquez, David Roh, Sachin Agarwal, Jan Claassen, E. Sander Connolly, Soojin Park^*

^*Corresponding author for this work

Neurology

Research output: Contribution to journal › Article › peer-review

7 Scopus citations

Abstract

Background: Prolonged external ventricular drainage (EVD) in patients with subarachnoid hemorrhage (SAH) leads to morbidity, whereas early removal can have untoward effects related to recurrent hydrocephalus. A metric to help determine the optimal time for EVD removal or ventriculoperitoneal shunt (VPS) placement would be beneficial in preventing the prolonged, unnecessary use of EVD. This study aimed to identify whether dynamics of cerebrospinal fluid (CSF) biometrics can temporally predict VPS dependency after SAH. Methods: This was a retrospective analysis of a prospective, single-center, observational study of patients with aneurysmal SAH who required EVD placement for hydrocephalus. Patients were divided into VPS-dependent (VPS+) and non–VPS dependent groups. We measured the bicaudate index (BCI) on all available computed tomography scans and calculated the change over time (ΔBCI). We analyzed the relationship of ΔBCI with CSF output by using Pearson’s correlation. A k-nearest neighbor model of the relationship between ΔBCI and CSF output was computed to classify VPS. Results: Fifty-eight patients met inclusion criteria. CSF output was significantly higher in the VPS+ group in the 7 days post EVD placement. There was a negative correlation between delta BCI and CSF output in the VPS+ group (negative delta BCI means ventricles become smaller) and a positive correlation in the VPS- group starting from days four to six after EVD placement (p < 0.05). A weighted k-nearest neighbor model for classification had a sensitivity of 0.75, a specificity of 0.70, and an area under the receiver operating characteristic curve of 0.80. Conclusions: The correlation of ΔBCI and CSF output is a reliable intraindividual biometric for VPS dependency after SAH as early as days four to six after EVD placement. Our machine learning model leverages this relationship between ΔBCI and cumulative CSF output to predict VPS dependency. Early knowledge of VPS dependency could be studied to reduce EVD duration in many centers (intensive care unit length of stay).

Original language	English (US)
Pages (from-to)	670-677
Number of pages	8
Journal	Neurocritical Care
Volume	37
Issue number	3
DOIs	https://doi.org/10.1007/s12028-022-01538-8
State	Published - Dec 2022

Funding

Dr. Park reports research support from the National Institutes of Health (R21 NS113055); Dr. Megjhani reports research support from the American Heart Association (20POST35210653); and Dr. Rubinos is supported by the University of North Carolina School of Medicine Physician Scientist Training Program.

Keywords

Cerebral spinal fluid dynamics
External ventricular drain
Hydrocephalus
Machine learning
Shunt dependency
Subarachnoid

ASJC Scopus subject areas

Critical Care and Intensive Care Medicine
Clinical Neurology

Access to Document

10.1007/s12028-022-01538-8

Cite this

Rubinos, C., Kwon, S. B., Megjhani, M., Terilli, K., Wong, B., Cespedes, L., Ford, J., Reyes, R., Kirsch, H., Alkhachroum, A., Velazquez, A., Roh, D., Agarwal, S., Claassen, J., Connolly, E. S., & Park, S. (2022). Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size. Neurocritical Care, 37(3), 670-677. https://doi.org/10.1007/s12028-022-01538-8

@article{ab51dfb367694c4fb06a7662310df09f,

title = "Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size",

abstract = "Background: Prolonged external ventricular drainage (EVD) in patients with subarachnoid hemorrhage (SAH) leads to morbidity, whereas early removal can have untoward effects related to recurrent hydrocephalus. A metric to help determine the optimal time for EVD removal or ventriculoperitoneal shunt (VPS) placement would be beneficial in preventing the prolonged, unnecessary use of EVD. This study aimed to identify whether dynamics of cerebrospinal fluid (CSF) biometrics can temporally predict VPS dependency after SAH. Methods: This was a retrospective analysis of a prospective, single-center, observational study of patients with aneurysmal SAH who required EVD placement for hydrocephalus. Patients were divided into VPS-dependent (VPS+) and non–VPS dependent groups. We measured the bicaudate index (BCI) on all available computed tomography scans and calculated the change over time (ΔBCI). We analyzed the relationship of ΔBCI with CSF output by using Pearson{\textquoteright}s correlation. A k-nearest neighbor model of the relationship between ΔBCI and CSF output was computed to classify VPS. Results: Fifty-eight patients met inclusion criteria. CSF output was significantly higher in the VPS+ group in the 7 days post EVD placement. There was a negative correlation between delta BCI and CSF output in the VPS+ group (negative delta BCI means ventricles become smaller) and a positive correlation in the VPS- group starting from days four to six after EVD placement (p < 0.05). A weighted k-nearest neighbor model for classification had a sensitivity of 0.75, a specificity of 0.70, and an area under the receiver operating characteristic curve of 0.80. Conclusions: The correlation of ΔBCI and CSF output is a reliable intraindividual biometric for VPS dependency after SAH as early as days four to six after EVD placement. Our machine learning model leverages this relationship between ΔBCI and cumulative CSF output to predict VPS dependency. Early knowledge of VPS dependency could be studied to reduce EVD duration in many centers (intensive care unit length of stay).",

keywords = "Cerebral spinal fluid dynamics, External ventricular drain, Hydrocephalus, Machine learning, Shunt dependency, Subarachnoid",

author = "Clio Rubinos and Kwon, {Soon Bin} and Murad Megjhani and Kalijah Terilli and Brenda Wong and Lizbeth Cespedes and Jenna Ford and Renz Reyes and Hannah Kirsch and Ayham Alkhachroum and Angela Velazquez and David Roh and Sachin Agarwal and Jan Claassen and Connolly, {E. Sander} and Soojin Park",

note = "Publisher Copyright: {\textcopyright} 2022, Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society.",

year = "2022",

month = dec,

doi = "10.1007/s12028-022-01538-8",

language = "English (US)",

volume = "37",

pages = "670--677",

journal = "Neurocritical Care",

issn = "1541-6933",

publisher = "Humana Press",

number = "3",

}

Rubinos, C, Kwon, SB, Megjhani, M, Terilli, K, Wong, B, Cespedes, L, Ford, J, Reyes, R, Kirsch, H, Alkhachroum, A, Velazquez, A, Roh, D, Agarwal, S, Claassen, J, Connolly, ES & Park, S 2022, 'Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size', Neurocritical Care, vol. 37, no. 3, pp. 670-677. https://doi.org/10.1007/s12028-022-01538-8

TY - JOUR

T1 - Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size

AU - Rubinos, Clio

AU - Kwon, Soon Bin

AU - Megjhani, Murad

AU - Terilli, Kalijah

AU - Wong, Brenda

AU - Cespedes, Lizbeth

AU - Ford, Jenna

AU - Reyes, Renz

AU - Kirsch, Hannah

AU - Alkhachroum, Ayham

AU - Velazquez, Angela

AU - Roh, David

AU - Agarwal, Sachin

AU - Claassen, Jan

AU - Connolly, E. Sander

AU - Park, Soojin

PY - 2022/12

Y1 - 2022/12

N2 - Background: Prolonged external ventricular drainage (EVD) in patients with subarachnoid hemorrhage (SAH) leads to morbidity, whereas early removal can have untoward effects related to recurrent hydrocephalus. A metric to help determine the optimal time for EVD removal or ventriculoperitoneal shunt (VPS) placement would be beneficial in preventing the prolonged, unnecessary use of EVD. This study aimed to identify whether dynamics of cerebrospinal fluid (CSF) biometrics can temporally predict VPS dependency after SAH. Methods: This was a retrospective analysis of a prospective, single-center, observational study of patients with aneurysmal SAH who required EVD placement for hydrocephalus. Patients were divided into VPS-dependent (VPS+) and non–VPS dependent groups. We measured the bicaudate index (BCI) on all available computed tomography scans and calculated the change over time (ΔBCI). We analyzed the relationship of ΔBCI with CSF output by using Pearson’s correlation. A k-nearest neighbor model of the relationship between ΔBCI and CSF output was computed to classify VPS. Results: Fifty-eight patients met inclusion criteria. CSF output was significantly higher in the VPS+ group in the 7 days post EVD placement. There was a negative correlation between delta BCI and CSF output in the VPS+ group (negative delta BCI means ventricles become smaller) and a positive correlation in the VPS- group starting from days four to six after EVD placement (p < 0.05). A weighted k-nearest neighbor model for classification had a sensitivity of 0.75, a specificity of 0.70, and an area under the receiver operating characteristic curve of 0.80. Conclusions: The correlation of ΔBCI and CSF output is a reliable intraindividual biometric for VPS dependency after SAH as early as days four to six after EVD placement. Our machine learning model leverages this relationship between ΔBCI and cumulative CSF output to predict VPS dependency. Early knowledge of VPS dependency could be studied to reduce EVD duration in many centers (intensive care unit length of stay).

AB - Background: Prolonged external ventricular drainage (EVD) in patients with subarachnoid hemorrhage (SAH) leads to morbidity, whereas early removal can have untoward effects related to recurrent hydrocephalus. A metric to help determine the optimal time for EVD removal or ventriculoperitoneal shunt (VPS) placement would be beneficial in preventing the prolonged, unnecessary use of EVD. This study aimed to identify whether dynamics of cerebrospinal fluid (CSF) biometrics can temporally predict VPS dependency after SAH. Methods: This was a retrospective analysis of a prospective, single-center, observational study of patients with aneurysmal SAH who required EVD placement for hydrocephalus. Patients were divided into VPS-dependent (VPS+) and non–VPS dependent groups. We measured the bicaudate index (BCI) on all available computed tomography scans and calculated the change over time (ΔBCI). We analyzed the relationship of ΔBCI with CSF output by using Pearson’s correlation. A k-nearest neighbor model of the relationship between ΔBCI and CSF output was computed to classify VPS. Results: Fifty-eight patients met inclusion criteria. CSF output was significantly higher in the VPS+ group in the 7 days post EVD placement. There was a negative correlation between delta BCI and CSF output in the VPS+ group (negative delta BCI means ventricles become smaller) and a positive correlation in the VPS- group starting from days four to six after EVD placement (p < 0.05). A weighted k-nearest neighbor model for classification had a sensitivity of 0.75, a specificity of 0.70, and an area under the receiver operating characteristic curve of 0.80. Conclusions: The correlation of ΔBCI and CSF output is a reliable intraindividual biometric for VPS dependency after SAH as early as days four to six after EVD placement. Our machine learning model leverages this relationship between ΔBCI and cumulative CSF output to predict VPS dependency. Early knowledge of VPS dependency could be studied to reduce EVD duration in many centers (intensive care unit length of stay).

KW - Cerebral spinal fluid dynamics

KW - External ventricular drain

KW - Hydrocephalus

KW - Machine learning

KW - Shunt dependency

KW - Subarachnoid

UR - http://www.scopus.com/inward/record.url?scp=85132789878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85132789878&partnerID=8YFLogxK

U2 - 10.1007/s12028-022-01538-8

DO - 10.1007/s12028-022-01538-8

M3 - Article

C2 - 35750930

AN - SCOPUS:85132789878

SN - 1541-6933

VL - 37

SP - 670

EP - 677

JO - Neurocritical Care

JF - Neurocritical Care

IS - 3

ER -

Predicting Shunt Dependency from the Effect of Cerebrospinal Fluid Drainage on Ventricular Size

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this