Validation of shunt dependency prediction scores after aneurysmal spontaneous subarachnoid hemorrhage

García-Armengol, Roser; Puyalto, Paloma; Misis, Maite; Rodríguez-Hernández, Ana; Julián, Joan Francesc; Pérez-Balaguero, Ana Cristina; Brugada-Bellsolà, Ferran; Menéndez, Belén; Cuadras, Patricia; Rimbau, Jordi

doi:10.1007/s00701-020-04688-w

Cited by 13 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It would be desirable to minimize the duration of external drainage and place a permanent shunt in all patients who need one as early as possible. To clarify this question, several prediction scores with the purpose of identifying exactly which patients eventually received a shunt have been published [ 8 , 9 , 12 , 14 , 24 ]. The prediction scores are typically based on clinical parameters such as Hunt Hess score, modified Fisher grade, and presence of acute hydrocephalus or an EVD [ 8 , 12 ].…”

Section: Discussionmentioning

confidence: 99%

Predictors of shunt insertion in patients with aneurysmal subarachnoid haemorrhage—a single-centre retrospective analysis

Capion,

Lilja-Cyron,

Olsen

et al. 2024

Acta Neurochir

View full text Add to dashboard Cite

Background No standard has been established regarding timing and choice of strategy for discontinuation of external ventricular drainage (EVD) in patients with aneurysmal subarachnoid haemorrhage (aSAH), and little is known about the importance of clinical variables. A proportion of the patients who initially pass their discontinuation attempt return with delayed hydrocephalus and the need of a permanent shunt. Early differentiation between patients who need a shunt and those who do not would facilitate care. We conducted a retrospective analysis on patients with aSAH and an EVD to search significant differences in treatment and clinical variables between patients who received a permanent shunt during initial hospitalization or after readmission, and those who never received a shunt. Methods We included 183 patients with aSAH who received an EVD over a 4-year period between 2015 and 2018 and divided them into three groups: those who received a shunt during primary admission, those who were readmitted for delayed hydrocephalus and received a shunt, and those who never needed a shunt. Between these groups, we compared selected clinical variables as well as outcome at discharge and after 6 months. Additionally, we assessed the ability of a shunt dependency score (SDASH) to predict the need for permanent drainage in the patients. Results Of 183 included patients, 108 (59%) ultimately received a ventriculoperitoneal (VP) shunt. Of these, 89 (82%) failed discontinuation during the primary admission and received a permanent shunt before discharge from the neurosurgical department. The remaining 19 (18%) were discharged after successful discontinuation, but subsequently developed delayed hydrocephalus and were admitted for shunt placement a median of 39 (range: 18–235) days after ictus. Ninety-four patients were discharged after successful discontinuation of the EVD, consisting of those who never developed the need for a permanent shunt and the 19 who were readmitted with delayed hydrocephalus, corresponding to a 20% (19/94) readmittance rate. Clinical variables such as drainage volume or discontinuation strategy did not differ across the three groups of patients. The SDASH score failed to provide any clinically useful information regarding prediction of shunt placement. Conclusion In this study, clinical variables including use of the predictive score SDASH predicted neither the overall need for nor the timing of shunt placement after aSAH. The homogeneous distribution of data between the three different groups renders strong independent clinical predictive factors unlikely. Thus, attempts to predict a permanent shunt requirement from these variables may be futile in these patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Predictors of shunt insertion in patients with aneurysmal subarachnoid haemorrhage—a single-centre retrospective analysis

Capion,

Lilja-Cyron,

Olsen

et al. 2024

Acta Neurochir

View full text Add to dashboard Cite

show abstract

“…The earlier published risk scores like the CHESS [ 27 ] and SDASH [ 14 ] scores were constructed for the prediction of shunt dependency in SAH and have been validated in other SAH cohorts [ 56 ]. Recently, Garcia-Armengol et al compared the Graeb, Hijdra [ 21 ], CHESS, and SDASH scores to predict shunt dependency in aneurysmal SAH patients [ 18 ]. While no significant differences were found for the SDASH and CHESS scores as well as the Graeb and Modified Graeb scores, the Hijdra score showed a lower predictive value.…”

Section: Discussionmentioning

confidence: 99%

The value of ventricular measurements in the prediction of shunt dependency after aneurysmal subarachnoid hemorrhage

et al. 2023

View full text Add to dashboard Cite

Objective Chronic hydrocephalus requiring shunt placement is a common complication of aneurysmal subarachnoid hemorrhage (SAH). Different risk factors and prediction scores for post-SAH shunt dependency have been evaluated so far. We analyzed the value of ventricle measurements for prediction of the need for shunt placement in SAH patients. Methods Eligible SAH cases treated between 01/2003 and 06/2016 were included. Initial computed tomography scans were reviewed to measure ventricle indices (bifrontal, bicaudate, Evans’, ventricular, Huckman’s, and third ventricle ratio). Previously introduced CHESS and SDASH scores for shunt dependency were calculated. Receiver operating characteristic analyses were performed for diagnostic accuracy of the ventricle indices and to identify the clinically relevant cut-offs. Results Shunt placement followed in 221 (36.5%) of 606 patients. In univariate analyses, all ventricular indices were associated with shunting (all: p<0.0001). The area under the curve (AUC) ranged between 0.622 and 0.662. In multivariate analyses, only Huckman’s index was associated with shunt dependency (cut-off at ≥6.0cm, p<0.0001) independent of the CHESS score as baseline prediction model. A combined score (0–10 points) containing the CHESS score components (0–8 points) and Huckman’s index (+2 points) showed better diagnostic accuracy (AUC=0.751) than the CHESS (AUC=0.713) and SDASH (AUC=0.693) scores and the highest overall model quality (0.71 vs. 0.65 and 0.67), respectively. Conclusions Ventricle measurements are feasible for early prediction of shunt placement after SAH. The combined prediction model containing the CHESS score and Huckman’s index showed remarkable diagnostic accuracy regarding identification of SAH individuals requiring shunt placement. External validation of the presented combined CHESS-Huckman score is mandatory.

show abstract

“…As a further feature, we used the amount of cerebrospinal fluid (CSF) drainage (in ml) over the first 14 days after aSAH, as it is an established risk factor of SDHC after aSAH and as it represents a “late feature” in a separate run (Fig. 1 ) [ 8 , 23 ]. Categorical features with at least three categories were transformed into binary features if they had too few instances per category.…”

Section: Methodsmentioning

confidence: 99%

“…Based on different clinical and radiological factors, various scoring systems have been developed to predict the risk of SDHC after aSAH [ 6 , 7 ]. The two validated scores with the best performance in SDHC prediction are the “Chronic Hydrocephalus Ensuing from SAH Score” (CHESS) and the “Shunt Dependency in aSAH Score” (SDASH) [ 2 , 8 , 9 ].…”

Section: Introductionmentioning

confidence: 99%

Enhancing the prediction for shunt-dependent hydrocephalus after aneurysmal subarachnoid hemorrhage using a machine learning approach

Frey,

Hilbert,

Früh

et al. 2023

Neurosurg Rev

View full text Add to dashboard Cite

Early and reliable prediction of shunt-dependent hydrocephalus (SDHC) after aneurysmal subarachnoid hemorrhage (aSAH) may decrease the duration of in-hospital stay and reduce the risk of catheter-associated meningitis. Machine learning (ML) may improve predictions of SDHC in comparison to traditional non-ML methods. ML models were trained for CHESS and SDASH and two combined individual feature sets with clinical, radiographic, and laboratory variables. Seven different algorithms were used including three types of generalized linear models (GLM) as well as a tree boosting (CatBoost) algorithm, a Naive Bayes (NB) classifier, and a multilayer perceptron (MLP) artificial neural net. The discrimination of the area under the curve (AUC) was classified (0.7 ≤ AUC < 0.8, acceptable; 0.8 ≤ AUC < 0.9, excellent; AUC ≥ 0.9, outstanding). Of the 292 patients included with aSAH, 28.8% (n = 84) developed SDHC. Non-ML-based prediction of SDHC produced an acceptable performance with AUC values of 0.77 (CHESS) and 0.78 (SDASH). Using combined feature sets with more complex variables included than those incorporated in the scores, the ML models NB and MLP reached excellent performances, with an AUC of 0.80, respectively. After adding the amount of CSF drained within the first 14 days as a late feature to ML-based prediction, excellent performances were reached in the MLP (AUC 0.81), NB (AUC 0.80), and tree boosting model (AUC 0.81). ML models may enable clinicians to reliably predict the risk of SDHC after aSAH based exclusively on admission data. Future ML models may help optimize the management of SDHC in aSAH by avoiding delays in clinical decision-making.

show abstract

Validation of shunt dependency prediction scores after aneurysmal spontaneous subarachnoid hemorrhage

Cited by 13 publications

References 25 publications

Predictors of shunt insertion in patients with aneurysmal subarachnoid haemorrhage—a single-centre retrospective analysis

Predictors of shunt insertion in patients with aneurysmal subarachnoid haemorrhage—a single-centre retrospective analysis

The value of ventricular measurements in the prediction of shunt dependency after aneurysmal subarachnoid hemorrhage

Enhancing the prediction for shunt-dependent hydrocephalus after aneurysmal subarachnoid hemorrhage using a machine learning approach

Contact Info

Product

Resources

About