A comparative study of fractional anisotropy measures and ICH score in predicting functional outcomes after intracerebral hemorrhage

Development and validation of a machine learning-based predictive model for assessing the 90-day prognostic outcome of patients with spontaneous intracerebral hemorrhage

BACKGROUND

Spontaneous intracerebral hemorrhage (sICH) is associated with significant mortality and morbidity. Predicting the prognosis of patients with sICH remains an important issue, which significantly affects treatment decisions. Utilizing readily available clinical parameters to anticipate the unfavorable prognosis of sICH patients holds notable clinical significance. This study employs five machine learning algorithms to establish a practical platform for the prediction of short-term prognostic outcomes in individuals afflicted with sICH.

METHODS

Within the framework of this retrospective analysis, the model underwent training utilizing data gleaned from 413 cases from the training center, with subsequent validation employing data from external validation center. Comprehensive clinical information, laboratory analysis results, and imaging features pertaining to sICH patients were harnessed as training features for machine learning. We developed and validated the model efficacy using all the selected features of the patients using five models: Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), XGboost and LightGBM, respectively. The process of Recursive Feature Elimination (RFE) was executed for optimal feature screening. An internal five-fold cross-validation was employed to pinpoint the most suitable hyperparameters for the model, while an external five-fold cross-validation was implemented to discern the machine learning model demonstrating the superior average performance. Finally, the machine learning model with the best average performance is selected as our final model while using it for external validation. Evaluation of the machine learning model's performance was comprehensively conducted through the utilization of the ROC curve, accuracy, and other relevant indicators. The SHAP diagram was utilized to elucidate the variable importance within the model, culminating in the amalgamation of the above metrics to discern the most succinct features and establish a practical prognostic prediction platform.

RESULTS

A total of 413 patients with sICH patients were collected in the training center, of which 180 were patients with poor prognosis. A total of 74 patients with sICH were collected in the external validation center, of which 26 were patients with poor prognosis. Within the training set, the test set AUC values for SVM, LR, RF, XGBoost, and LightGBM models were recorded as 0.87, 0.896, 0.916, 0.885, and 0.912, respectively. The best average performance of the machine learning models in the training set was the RF model (average AUC: 0.906 ± 0.029, P < 0.01). The model still maintains a good performance in the external validation center, with an AUC of 0.817 (95% CI 0.705-0.928). Pertaining to feature importance for short-term prognostic attributes of sICH patients, the NIHSS score reigned supreme, succeeded by AST, Age, white blood cell, and hematoma volume, among others. In culmination, guided by the RF model's variable importance weight and the model's ROC curve insights, the NIHSS score, AST, Age, white blood cell, and hematoma volume were integrated to forge a short-term prognostic prediction platform tailored for sICH patients.

CONCLUSION

We constructed a prediction model based on the results of the RF model incorporating five clinically accessible predictors with reliable predictive efficacy for the short-term prognosis of sICH patients. Meanwhile, the performance of the external validation set was also more stable, which can be used for accurate prediction of short-term prognosis of sICH patients.

Acute corticospinal tract diffusion tensor imaging predicts 6-month functional outcome after intracerebral haemorrhage

INTRODUCTION

Diffusion tensor imaging (DTI) can assess the structural integrity of the corticospinal tract (CST) in vivo. We aimed to investigate whether CST DTI metrics after intracerebral haemorrhage (ICH) are associated with 6-month functional outcome and can improve the predictive performance of the existing ICH score.

METHODS

We retrospectively included 42 patients with DTI performed within 5 days after deep supratentorial spontaneous ICH. Ipsilesional-to-contralesional ratios were calculated for fractional anisotropy (rFA) and mean diffusivity (rMD) in the pontine segment (PS) of the CST. We determined the most predictive variables for poor 6-month functional outcome [modified Rankin Scale (mRS) > 2] using the least absolute shrinkage and selection operator (LASSO) method. We calculated discrimination using optimism-adjusted estimation of the area under the curve (AUC).

RESULTS

Patients with 6-month mRS > 2 had lower rFA (0.945 [± 0.139] vs 1.045 [± 0.130]; OR 0.004 [95% CI 0.00-0.77]; p =  0.04) and higher rMD (1.233 [± 0.418] vs 0.963 [± 0.211]; OR 22.5 [95% CI 1.46-519.68]; p = 0.02). Discrimination (AUC) values were: 0.76 (95% CI 0.61-0.91) for the ICH score, 0.71 (95% CI 0.54-0.89) for rFA, and 0.72 (95% CI 0.61-0.91) for rMD. Combined models with DTI and non-DTI variables offer an improvement in discrimination: for the best model, the AUC was 0.82 ([95% CI 0.68-0.95]; p = 0.15).

CONCLUSION

In our exploratory study, PS-CST rFA and rMD had comparable predictive ability to the ICH score for 6-month functional outcome. Adding DTI metrics to clinical-radiological scores might improve discrimination, but this needs to be investigated in larger studies.

Diffusion Tensor Imaging as a Prognostic Tool for Recovery in Acute and Hyperacute Stroke

Stroke represents a major cause of mortality and long-term disability among adult populations, leaving a devastating socioeconomic impact globally. Clinical manifestation of stroke is characterized by great diversity, ranging from minor disability to considerable neurological impairment interfering with activities of daily living and even death. Prognostic ambiguity has stimulated the interest for implementing stroke recovery biomarkers, including those provided by structural neuroimaging techniques, i.e., diffusion tensor imaging (DTI) and tractography for the study of white matter (WM) integrity. Considering the necessity of prompt and accurate prognosis in stroke survivors along with the potential capacity of DTI as a relevant imaging biomarker, the purpose of our study was to review the pertinent literature published within the last decade regarding DTI as a prognostic tool for recovery in acute and hyperacute stroke. We conducted a thorough literature search in two databases (MEDLINE and Science Direct) in order to trace all relevant studies published between 1 January 2012 and 16 March 2022 using predefined terms as key words. Only full-text human studies published in the English language were included. Forty-four studies were identified and are included in this review. We present main findings and by describing several methodological issues, we highlight shortcomings and gaps in the current literature so that research priorities for future research can be outlined. Our review suggests that DTI can track longitudinal changes and identify prognostic correlates in acute and hyperacute stroke patients.

Scoping Review and Commentary on Prognostication for Patients with Intracerebral Hemorrhage with Advances in Surgical Techniques

INTRODUCTION

The intracerebral hemorrhage (ICH) score provides an estimate of 30-day mortality for patients with intracerebral hemorrhage in order to guide research protocols and clinical decision making. Several variations of such scoring systems have attempted to optimize its prognostic value. More recently, minimally invasive surgical techniques are increasingly being used with promising results. As more patients become candidates for surgical intervention, there is a need to re-discuss the best methods for predicting outcomes with or without surgical intervention.

METHODS

We systematically performed a scoping review with a comprehensive literature search by two independent reviewers using the PubMed and Cochrane databases for articles pertaining to the "intracerebral hemorrhage score." Relevant articles were selected for analysis and discussion of potential modifications to account for increasing surgical indications.

RESULTS

A total of 64 articles were reviewed in depth and identified 37 clinical grading scales for prognostication of spontaneous intracerebral hemorrhage. The original ICH score remains the most widely used and validated. Various authors proposed modifications for improved prognostic accuracy, though no single scale showed consistent superiority. Most recently, scales to account for advances in surgical techniques have been developed but lack external validation.

CONCLUSION

We provide the most comprehensive review to date of prognostic grading scales for patients with intracerebral hemorrhage. Current prognostic tools for patients with intracerebral hemorrhage remain limited and may overestimate risk of a poor outcome. As minimally invasive surgical techniques are developed, prognostic scales should account for surgical candidacy and outcomes.

White Matter Survival within and around the Hematoma: Quantification by MRI in Patients with Intracerebral Hemorrhage

White matter (WM) injury and survival after intracerebral hemorrhage (ICH) has received insufficient attention. WM disruption surrounding the hematoma has been documented in animal models with histology, but rarely in human ICH with noninvasive means, like magnetic resonance imaging (MRI). A few human MRI studies have investigated changes in long WM tracts after ICH remote from the hematoma, like the corticospinal tract, but have not attempted to obtain an unbiased quantification of WM changes within and around the hematoma over time. This study attempts such quantification from 3 to 30 days post ictus. Thirteen patients with mild to moderate ICH underwent diffusion tensor imaging (DTI) MRI at 3, 14, and 30 days. Fractional anisotropy (FA) maps were used to calculate the volume of tissue with FA > 0.5, both within the hematoma (lesion) and in the perilesional tissue. At day 3, the percentages of both lesional and perilesional tissue with an FA > 0.5 were significantly less than contralateral, unaffected, anatomically identical tissue. This perilesional contralateral difference persisted at day 14, but there was no significant difference at day 30. The loss of perilesional tissue with FA > 0.5 increased with increasing hematoma size at day 3 and day 14. All patients had some tissue within the lesion with FA > 0.5 at all time points. This did not decrease with duration after ictus, suggesting the persistence of white matter within the hematoma/lesion. These results outline an approach to quantify WM injury, both within and surrounding the hematoma, after mild to moderate ICH using DTI MRI. This may be important for monitoring treatment strategies, such as hematoma evacuation, and assessing efficacy noninvasively.

Cerebral Small Vessel Disease and Functional Outcome Prediction After Intracerebral Hemorrhage

OBJECTIVE

To determine whether CT-based cerebral small vessel disease (SVD) biomarkers are associated with 6-month functional outcome after intracerebral hemorrhage (ICH) and whether these biomarkers improve the performance of the preexisting ICH prediction score.

METHODS

We included 864 patients with acute ICH from a multicenter, hospital-based prospective cohort study. We evaluated CT-based SVD biomarkers (white matter hypodensities [WMH], lacunes, brain atrophy, and a composite SVD burden score) and their associations with poor 6-month functional outcome (modified Rankin Scale score >2). The area under the receiver operating characteristic curve (AUROC) and Hosmer-Lemeshow test were used to assess discrimination and calibration of the ICH score with and without SVD biomarkers.

RESULTS

In multivariable models (adjusted for ICH score components), WMH presence (odds ratio [OR] 1.52, 95% confidence interval [CI] 1.12-2.06), cortical atrophy presence (OR 1.80, 95% CI 1.19-2.73), deep atrophy presence (OR 1.66, 95% CI 1.17-2.34), and severe atrophy (either deep or cortical) (OR 1.94, 95% CI 1.36-2.74) were independently associated with poor functional outcome. For the revised ICH score, the AUROC was 0.71 (95% CI 0.68-0.74). Adding SVD markers did not significantly improve ICH score discrimination; for the best model (adding severe atrophy), the AUROC was 0.73 (95% CI 0.69-0.76). These results were confirmed when lobar and nonlobar ICH were considered separately.

CONCLUSIONS

The ICH score has acceptable discrimination for predicting 6-month functional outcome after ICH. CT biomarkers of SVD are associated with functional outcome, but adding them does not significantly improve ICH score discrimination.

TRIAL REGISTRATION INFORMATION

ClinicalTrials.gov Identifier: NCT02513316.

Peri-hematoma corticospinal tract integrity in intracerebral hemorrhage patients: A diffusion-tensor imaging study

BACKGROUND

The impact of perihematoma edema in Intracerebral Hemorrhage (ICH) on white matter integrity is uncertain. Fractional Anisotropy (FA), as measured with Diffusion Tensor Imaging (DTI), can be used to assess white matter microstructure. We tested the hypotheses that sections of the Corticospinal Tract (CST) passing through perihematoma edema would 1) have low FA relative to the contralateral CST and 2) would predict NIHSS motor score in ICH patients.

METHODS

Patients were prospectively imaged with DTI at 48 h and 7 days after onset. Edema volume/extent was measured on CT at baseline and 24 h. FA, mean, axial and radial diffusivity were measured in the perihematoma edema, contralateral CST and sections of CST passing through the edema ('edematous CST').

RESULTS

Patients (n = 27, mean age 67 ± 13) were scanned with DTI at a median (IQR) of 42.3 (24.5) hours and 7.7 (1.8) days from onset. Median acute ICH volume was 8.8 (22) ml. FA in edematous CST at 72 h was decreased (0.37 ± 0.03) relative to contralateral CST (0.52 ± 0.06; p < 0.0001). Day 7 FA in edematous CST (0.35 ± 0.08) was also decreased compared to contralateral CST (0.54 ± 0.06; p < 0.0001). FA remained stable between 72 h (0.37 ± 0.03) and day 7 (0.35 ± 0.07; p = 0.350). FA at 72 h (ρ = -0.22, p = 0.420) and day 7 (ρ = -0.14, p = 0.624) was unrelated to 90-day motor score.

CONCLUSIONS

FA is decreased in the CST where it passes through the edema. Decreased FA in the edematous CST remained stable over time, was unrelated to motor score, and may represent water infiltration into the tracts rather than axonal injury.

Evidence of motor injury due to damaged corticospinal tract following acute hemorrhage in the basal ganglia region

The integrity of the corticospinal tract (CST) is significantly affected following basal ganglia haemorrhage. We aimed to assess the local features of CST and to effectively predict motor function by diffusion characteristics of CST in patients with motor injury following acute haemorrhage in the acute basal ganglia region. We recruited 37 patients with paresis of the lateral limbs caused by acute basal ganglia haemorrhage. Based on the automated fiber quantification method to track CST, assessed the character of each CST segment between the affected and contralateral sides, and correlated these with the Fugl-Meyer (FM) and Barthel Index (BI) scores at 6 months after onset. The fractional anisotropy (FA) values of the injured side of CST showed a significantly lower FA than the contralateral side along the tract profiles (p < 0.05, corrections for multiple comparisons). The FA values of each site at the internal capsule, closed corona radiata were positively correlated with the FM and BI score at 6 months after onset (p < 0.001, respectively). Our findings assessed the character of CST vividly in detail and dementated the primary sites of CST can predict the long-term outcome of motor function. This study may facilitate future clinical and cognitive studies of acute haemorrhage.

Quantitative Corticospinal Tract Assessment in Acute Intracerebral Hemorrhage

Intracerebral hemorrhage (ICH) prognostication during the acute phase is often subjective among physicians and often affects treatment decisions. The present study explores objective imaging parameters using quantitative corticospinal tract (CST) fiber reconstruction during the acute phase of ICH and correlates these parameters with functional outcome and patient recovery. We prospectively enrolled nonsurgical spontaneous supratentorial ICH patients and obtained an MRI scan on day 5 ± 1. Q-space diffeomorphic reconstruction was performed using DSI Studio, and quantitative anisotropy (QA) was calculated. The CST was reconstructed based on QA. The dichotomized modified Rankin Scale score on day 90 (favorable outcome = 0–2) and Barthel Index (favorable recovery = 100 on day 90 or improvement between discharge and day 90 > 60%) were assessed. Thirty-three patients, median age 72 years (interquartile range (IQR) 64–83), 21 female (64%), 21 (64%) with lobar hemorrhage, median ICH volume on admission 15.0 (IQR 7.0–27.4) mL, were included. Sixteen patients (48%) had a favorable outcome and 24 (73%) had a favorable recovery. The mean number of ipsilesional reconstructed CST fiber pathways was higher in patients with favorable outcomes (153 (standard deviation (SD) 103) vs. 60 (SD 39), p = 0.003) and predicted outcome after adjustment (Exp(B) = 1.016 (95% CI = 1.002–1.030)). QA in the ipsilesional posterior limb of the internal capsule showed a trend towards an association with favorable outcome (Exp(B) = 1.194 (95% CI = 0.991–1.439 (adjusted))). The total (ipsilesional + contralesional) number of reconstructed fiber pathways was associated with favorable recovery (Exp(B) = 1.025 (95% CI = 1.003–1.047 (adjusted))). Quantitative tractography parameters assessed in the acute phase of ICH may represent a promising predictor of long-term outcome and recovery. This might facilitate prognostic evaluation and organization of rehabilitation.

Assessment and Comparison of the Four Most Extensively Validated Prognostic Scales for Intracerebral Hemorrhage: Systematic Review with Meta-analysis

BACKGROUND/OBJECTIVE

Intracerebral hemorrhage (ICH) is a devastating disorder, responsible for 10% of all strokes. Several prognostic scores have been developed for this population to predict mortality and functional outcome. The aim of this study was to determine the four most frequently validated and most widely used scores, assess their discrimination for both outcomes by means of a systematic review with meta-analysis, and compare them using meta-regression.

METHODS

PubMed, ISI Web of Knowledge, Scopus, and CENTRAL were searched for studies validating the ICH score, ICH-GS, modified ICH, and the FUNC score in ICH patients. C-statistic was chosen as the measure of discrimination. For each score and outcome, C-statistics were aggregated at four different time points using random effect models, and heterogeneity was evaluated using the I² statistic. Score comparison was undertaken by pooling all C-statistics at different time points using robust variance estimation (RVE) and performing meta-regression, with the score used as the independent variable.

RESULTS

Fifty-three studies were found validating the original ICH score, 14 studies were found validating the ICH-GS, eight studies were found validating the FUNC score, and five studies were found validating the modified ICH score. Most studies attempted outcome prediction at 3 months or earlier. Pooled C-statistics ranged from 0.76 for FUNC functional outcome prediction at discharge to 0.85 for ICH-GS mortality prediction at 3 months, but heterogeneity was high across studies. RVE showed the ICH score retained the highest discrimination for mortality (c = 0.84), whereas the modified ICH score retained the highest discrimination for functional outcome (c = 0.80), but these differences were not statistically significant.

CONCLUSIONS

The ICH score is the most extensively validated score in ICH patients and, in the absence of superior prediction by other scores, should preferably be used. Further studies are needed to validate prognostic scores at longer follow-ups and assess the reasons for heterogeneity in discrimination.

Predicting Motor Outcome in Acute Intracerebral Hemorrhage

BACKGROUND AND PURPOSE

Predicting motor outcome following intracerebral hemorrhage is challenging. We tested whether the combination of clinical scores and DTI-based assessment of corticospinal tract damage within the first 12 hours of symptom onset after intracerebral hemorrhage predicts motor outcome at 3 months.

MATERIALS AND METHODS

We prospectively studied patients with motor deficits secondary to primary intracerebral hemorrhage within the first 12 hours of symptom onset. Patients underwent multimodal MR imaging including DTI. We assessed intracerebral hemorrhage and perihematomal edema location and volume, and corticospinal tract involvement. The corticospinal tract was considered affected when the tractogram passed through the intracerebral hemorrhage or/and the perihematomal edema. We also calculated affected corticospinal tract-to-unaffected corticospinal tract ratios for fractional anisotropy, mean diffusivity, and axial and radial diffusivities. Motor impairment was graded by the motor subindex scores of the modified NIHSS. Motor outcome at 3 months was classified as good (modified NIHSS 0-3) or poor (modified NIHSS 4-8).

RESULTS

Of 62 patients, 43 were included. At admission, the median NIHSS score was 13 (interquartile range = 8-17), and the median modified NIHSS score was 5 (interquartile range = 2-8). At 3 months, 13 (30.23%) had poor motor outcome. Significant independent predictors of motor outcome were NIHSS and modified NIHSS at admission, posterior limb of the internal capsule involvement by intracerebral hemorrhage at admission, intracerebral hemorrhage volume at admission, 72-hour NIHSS, and 72-hour modified NIHSS. The sensitivity, specificity, and positive and negative predictive values for poor motor outcome at 3 months by a combined modified NIHSS of >6 and posterior limb of the internal capsule involvement in the first 12 hours from symptom onset were 84%, 79%, 65%, and 92%, respectively (area under the curve = 0.89; 95% CI, 0.78-1).

CONCLUSIONS

Combined assessment of motor function and posterior limb of the internal capsule damage during acute intracerebral hemorrhage accurately predicts motor outcome.

Intracerebral Hemorrhage Location and Functional Outcomes of Patients: A Systematic Literature Review and Meta-Analysis

BACKGROUND AND PURPOSE

Intracerebral hemorrhage (ICH) has the highest mortality rate among all strokes. While ICH location, lobar versus non-lobar, has been established as a predictor of mortality, less is known regarding the relationship between more specific ICH locations and functional outcome. This review summarizes current work studying how ICH location affects outcome, with an emphasis on how studies designate regions of interest.

METHODS

A systematic search of the OVID database for relevant studies was conducted during August 2015. Studies containing an analysis of functional outcome by ICH location or laterality were included. As permitted, the effect size of individual studies was standardized within a meta-analysis.

RESULTS

Thirty-seven studies met the inclusion criteria, the majority of which followed outcome at 3 months. Most studies found better outcomes on the Modified Rankin Scale (mRS) or Glasgow Outcome Score (GOS) with lobar compared to deep ICHs. While most aggregated deep structures for analysis, some studies found poorer outcomes for thalamic ICH in particular. Over half of the studies did not have specific methodological considerations for location designations, including blinding or validation.

CONCLUSIONS

Multiple studies have examined motor-centric outcomes, with few studies examining quality of life (QoL) or cognition. Better functional outcomes have been suggested for lobar versus non-lobar ICH; few studies attempted finer topographic comparisons. This study highlights the need for improved reporting in ICH outcomes research, including a detailed description of hemorrhage location, reporting of the full range of functional outcome scales, and inclusion of cognitive and QoL outcomes.

Can Neurological Biomarkers of Brain Impairment Be Used to Predict Poststroke Motor Recovery? A Systematic Review

Background There is growing interest to establish recovery biomarkers, especially neurological biomarkers, in order to develop new therapies and prediction models for the promotion of stroke rehabilitation and recovery. However, there is no consensus among the neurorehabilitation community about which biomarker(s) have the highest predictive value for motor recovery. Objective To review the evidence and determine which neurological biomarker(s) meet the high evidence quality criteria for use in predicting motor recovery. Methods We searched databases for prognostic neuroimaging/neurophysiological studies. Methodological quality of each study was assessed using a previously employed comprehensive 15-item rating system. Furthermore, we used the GRADE approach and ranked the overall evidence quality for each category of neurologic biomarker. Results Seventy-one articles met our inclusion criteria; 5 categories of neurologic biomarkers were identified: diffusion tensor imaging (DTI), transcranial magnetic stimulation (TMS), functional magnetic resonance imaging (fMRI), conventional structural MRI (sMRI), and a combination of these biomarkers. Most studies were conducted with individuals after ischemic stroke in the acute and/or subacute stage (~70%). Less than one-third of the studies (21/71) were assessed with satisfactory methodological quality (80% or more of total quality score). Conventional structural MRI and the combination biomarker categories ranked "high" in overall evidence quality. Conclusions There were 3 prevalent methodological limitations: (a) lack of cross-validation, (b) lack of minimal clinically important difference (MCID) for motor outcomes, and (c) small sample size. More high-quality studies are needed to establish which neurological biomarkers are the best predictors of motor recovery after stroke. Finally, the quarter-century old methodological quality tool used here should be updated by inclusion of more contemporary methods and statistical approaches.

Prediction of upper extremity motor recovery after subacute intracerebral hemorrhage through diffusion tensor imaging: a systematic review and meta-analysis

INTRODUCTION

Early assessment of the pyramidal tracts is important for intracerebral hemorrhage (ICH) patients in order to decide the optimal treatment or to assess appropriate rehabilitation strategies, and management of patient expectations and goals. The purpose of this study was to systematically review and summarize the current available literature on the value of Fractional Anisotropy (FA) parameter of the diffusion tensor imaging (DTI) in predicting upper extremity (UE) motor recovery after subacute ICH.

METHODS

PubMed, EMBASE, MEDLINE, Google Scholar, and Cochrane CENTRAL searches were conducted from 1 January 1950 to 31 March 2016 which were supplemented with relevant articles identified in the references. Pooled estimate using correlation between DTI parameter FA and UE motor recovery was done using comprehensive meta-analysis software.

RESULTS

Out of 97 citations, only eight studies met the criteria for inclusion in the systematic review and six studies were included in the meta-analysis. A random effects model revealed that DTI parameter FA is a significant predictor for UE motor recovery after subacute ICH (correlation coefficient = 0.56; 95 % confidence interval 0.44 to 0.65, P value <0.001). However, moderate heterogeneity was observed between the studies (Tau-squared = 0.28, I-squared = 70.3).

CONCLUSION

The studies reported so far on correlation between FA parameter of DTI and UE motor recovery in ICH patients are few with small sample sizes. This meta-analysis suggests a strong correlation between DTI parameter FA and UE motor recovery in ICH patients. Further well-designed prospective studies embedded with larger sample size are needed to confirm these findings.