Elsevier

Journal of the Neurological Sciences

Volume 396, 15 January 2019, Pages 219-224
Journal of the Neurological Sciences

Hypodensities detected at 1.5–3 h after intracerebral hemorrhage better predicts secondary neurological deterioration

https://doi.org/10.1016/j.jns.2018.11.032Get rights and content

Highlights

  • There has been no study working on the optimal detecting time of a certain NCCT imaging feature.

  • Hypodensities within hematoma detected by an NCCT scan may predict secondary neurological deterioration.

  • Hypodensities detected at 1.5–3.0 h after ICH onset have better predictive efficacy.

Abstract

Background

Secondary neurological deterioration in patients with spontaneous intracerebral hemorrhage (sICH) develops within the first 24 or 48 h after ICH onset and appears to portend a poor prognosis. We aimed to verify whether hypodensities within an acute ICH detected by noncontrast computed tomography (NCCT) were able to predict secondary neurological deterioration and investigate which monitoring window was of the highest predictive value.

Materials/methods

This study involved sICH patients from three clinical centers of Fudan University between October 1, 2016 and March 31, 2018. Logistic regression analysis was used to assess the association between hypodensities and secondary neurological deterioration. The receiver operating characteristic curve of the subjects was performed to evaluate the critical value of the detection time window of hypodensities that best predicted the secondary neurological injury. Then, we divided the detection time window of hypodensities into 0–1.5 h, 1.5–3 h, 3–4.5 h and 4.5–6 h, and calculated the sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV) and accuracy respectively.

Results

A total of 240 ICH patients met the inclusion criteria, 97 (40.42%) of whom were observed secondary neurological deterioration. Hypodensities were positive in 113 patients (47.08%), and more common in patients with secondary neurological deterioration (76.25%). The multivariate logistic regression analysis demonstrated that infratentorial hemorrhage (P < .001), the baseline hematoma volume (P = .015), and the presence of hypodensities on admission CT scan (P < .001) were independent predictors of secondary neurological deterioration. The sensitivity, specificity, PPV, and NPV of hypodensities in predicting secondary neurological deterioration were 76.3%, 72.7%, 65.5%, and 81.9%, respectively. When the time to the baseline NCCT was 114.5 min, the hypodensities were of the highest predictive value. Besides, the risk of secondary neurological deterioration predicted by hypodensities detected during 1.5–3.0 h was higher than other time periods.

Conclusions

Hypodensities within hematoma detected by an NCCT scan may predict secondary neurological deterioration, independent of other clinical and imaging predictors. Hypodensities detected at 1.5–3.0 h after ICH onset have better predictive efficacy.

Introduction

Spontaneous intracerebral hemorrhage (sICH) is a severe type of stroke with high morbidity and mortality throughout the world, which accounts for approximately 15% of all strokes worldwide [[1], [2], [3]]. Several studies have investigated early neurological deterioration in patients with ICH, which develops within the first 24 or 48 h after symptom onset, and appears to portend a poor prognosis [4].

Hematoma expansion, which often occurs within 6 h after ICH onset, is closely associated with early deterioration and poor outcomes [5]. The definition of hematoma expansion varies in different literature. Some studies define hematoma expansion as a proportional increase of 33% or an absolute increase of >6 mL [6]. In addition, it is also defined as an increase of hematoma volume > 33% or absolute >12.5 mL from initial CT scan in the large clinical trial such as INTERACT 1 and rFVIIa ICH trail [3,7]. The computed tomography angiography (CTA) spot sign is a well-established imaging marker that independently predicts secondary neurological deterioration in patients with sICH [8]. However, early CTA examination requires contrast administration and is contraindicated in patients with several renal function impairment [9]. Therefore, the markers detected by noncontrast CT (NCCT) scan as possible imaging predictors for secondary neurological deterioration are identified. The swirl sign is described as region(s) of hypoattenuation or isoattenuation (compared to the attenuation of brain parenchyma) within the hyperattenuated ICH [10]. But actually, swirl sign is not well-defined in a global context, so it is hard to make a unified definition in different medical institutions. The blend sign on NCCT, which represents the blending of the hypoattenuating area and the hyperattenuating region with a clear margin, has been introduced as a predictor for hematoma expansion [11]. Blend sign can be used to predict secondary neurological deterioration as well [7]. Nevertheless, it also has drawbacks such as hyposensitivity. Meanwhile, the definition based on CT attenuation values (Hounsfield units, HU) makes it difficult to identify just from a single CT film. Hypodensities, which have been accepted as reliable tools for predicting hematoma expansion in ICH patients [12], are much easier to identify in NCCT. In addition, it has been hinted that hypodensities are associated with 90-day poor outcome in patients with sICH [13].

To predict secondary neurological deterioration and offer the patients proper treatments in time, we aimed to investigate the relationship between hypodensities and secondary neurological deterioration. Moreover, we also made a comprehensive evaluation of the best predictive time period at which hypodensities were detected.

Section snippets

Materials and methods

This study involved sICH patients aged 18 years or older between October 1, 2016 and March 31, 2018 from three clinical centers of Fudan University: Huanshan Hospital, Zhongshan Hospital and Jinshan Hospital. In this prospective study, patients were eligible for the study providing the baseline NCCT scan was performed within 6 h after sICH onset. The exclusion criteria included secondary ICH (cerebral aneurysm, moyamoya syndrome, arteriovenous malformation, tumor, or hemorrhagic transformation

Baseline characteristics

43 of 283 patients with primary ICH were excluded (time to baseline NCCT>6 h: 7; inclusion disapproval: 13; secondary ICH: 6; primary IVH: 8; and baseline ICH volume < 1 ml: 9). A total of 240 patients (male: 179, female: 61) met the inclusion criteria, 97 (40.42%) of whom were found secondary neurological deterioration. The mean age of the patients was 62.9 ± 11.5 years (age range, 25–90 years). The median (IQR) of baseline hematoma volume was 22.25(4.43–27.78) ml. The median (IQR) of time to

Discussion

sICH remains a serious condition with a 30-day mortality of approximately 40%, half of which occurs within the first 48 h [15]. Patients with sICH may have any level of neurological deficit at the time of hematoma formation, ranging from none through to complete motor and sensory paraplegia. As to secondary neurological deterioration, we defined [1] early hemi-craniectomy under standardized criteria according to our in-house guidelines or [2] secondary decrease of GCS of >3 points, both within

Conclusions

Hypodensities within hematoma detected by an NCCT scan may predict secondary neurological deterioration, independent of other clinical and imaging predictors. Hypodensities detected at 1.5–3.0 h after ICH onset have better predictive efficacy.

Funding

This research was supported by grants from the National Natural Science Foundation of China (grant number 81500998); and Science and Technology Commission of Shanghai Municipality (grant number 16140903200).

Competing interests

None declared.

References (24)

  • Q. Li et al.

    Black Hole Sign: Novel imaging marker that predicts hematoma growth in patients with intracerebral hemorrhage

    Stroke

    (2016)
  • E. Selariu et al.

    Swirl sign in intracerebral haemorrhage: definition, prevalence, reliability and prognostic value

    BMC Neurol.

    (2012)
  • Cited by (0)

    1

    These authors contributed equally to this work.

    View full text