Prediction of clinical outcome in subacute subarachnoid hemorrhage using diffusion tensor imaging

Restricted access

OBJECTIVE

Clinical outcome in nontraumatic subarachnoid hemorrhage (SAH) is multifactorial and difficult to predict. Diffusion tensor imaging (DTI) findings are a prognostic marker in some diseases such as traumatic brain injury. The authors hypothesized that DTI parameters measured in the subacute phase of SAH can be associated with a poor clinical outcome.

METHODS

Diffusion tensor imaging was prospectively performed in 54 patients at 8–10 days after nontraumatic SAH. Logistic regression analysis was performed to evaluate the association of fractional anisotropy (FA) and apparent diffusion coefficient (ADC) values with a poor clinical outcome (modified Rankin Scale score ≥ 3) at 3 months.

RESULTS

At 8–10 days post-SAH, after adjusting for other variables associated with a poor outcome, an increased ADC at the frontal centrum semiovale was associated with a poor prognosis (OR estimate 1.29, 95% CI 1.04–1.60, p = 0.020). Moreover, an increase of 0.1 in the FA value at the corpus callosum at 8–10 days after SAH corresponded to 66% lower odds of having a poor outcome (p = 0.002).

CONCLUSIONS

Decreased FA and increased ADC values in specific brain regions were independently associated with a poor clinical outcome after SAH. This preliminary exploratory study supports a potential role for DTI in predicting the outcome of SAH.

ABBREVIATIONS ADC = apparent diffusion coefficient; AUC = area under the ROC curve; DCI = delayed cerebral ischemia; DTI = diffusion tensor imaging; DWI = diffusion-weighted imaging; FA = fractional anisotropy; GCS = Glasgow Coma Scale; HH = Hunt and Hess; MRI = magnetic resonance imaging; mRS = modified Rankin Scale; PLIC = posterior limb of internal capsule; ROC = receiver operating characteristic; ROI = region of interest; SAH = subarachnoid hemorrhage; WFNS = World Federation of Neurosurgical Societies.

Article Information

Correspondence Isabel Fragata: Centro Hospitalar Lisboa Central, Lisbon, Portugal. isabelfragata@gmail.com.

INCLUDE WHEN CITING Published online April 13, 2018; DOI: 10.3171/2017.10.JNS171793.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Transverse FA and ADC maps show ROIs drawn bilaterally in the brain parenchyma. Upper and lower left: Regions of interest in the pons and cerebellar white matter. Upper and lower center: Regions of interest in the lentiform nucleus, thalamus, PLIC, and splenium and genu of corpus callosum. Upper and lower right: Regions of interest in the frontal and parietal centrum semiovale. From Neuroradiology, Evolution of diffusion tensor imaging parameters after acute subarachnoid haemorrhage: a prospective cohort study, 59, 2017, 13–21, Fragata I, Canhão P, Alves M, Papoila AL, Canto-Moreira N. With permission from Springer. Figure is available in color online only.

  • View in gallery

    Flowchart with the number of patients admitted, excluded, and enrolled in the present study. h = hours; IV = intravenous.

  • View in gallery

    Influence of age on outcome at 3 months (solid curve) and corresponding 95% confidence intervals (dashed curves), showing a nonlinear association that led to a dichotomization into 2 categories: age 42–67 years and ages < 42/> 67 years.

  • View in gallery

    Odds ratio estimates and corresponding 95% confidence intervals obtained by logistic regression models for a poor clinical outcome in each ROI. A: Fractional anisotropy at 8–10 days post-SAH. B: Apparent diffusion coefficient at 8–10 days post-SAH. C = cerebellum; FCSO = frontal centrum semiovale; GSCC = genu and splenium of corpus callosum; L = lentiform nucleus; P = pons; PCSO = parietal centrum semiovale; T = thalamus.

References

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 198 198 23
Full Text Views 73 73 11
PDF Downloads 86 86 7
EPUB Downloads 0 0 0

PubMed

Google Scholar