We analyzed the clinical and imaging data of patients with supratentorial sICH who underwent stereotactic aspiration and catheter drainage, aiming to identify factors conducive to HE and improve the effect of minimally invasive surgery.
The presence of a positive NCCT blend sign prior to surgery significantly enhances HE on both the first postoperative day and before tube removal. Conversely, a favorable catheter position score only promotes HE on the initial postoperative day, exhibiting no significant correlation with HE at the later time point. Furthermore, our analysis demonstrates that a larger hematoma volume is deleterious to HE on the first postoperative day, with an optimal cutoff value identified as 45.46ml. Notably, the presence of irregular hematoma morphology and a hematoma edge contiguous with the ventricle are adverse factors for HE prior to extubation (Fig. 3).
Previous scholarly investigations have demonstrated that the blend sign can serve as an indicator of active bleeding and reflect various stages of hematoma development [24]. Notably, the attenuation values observed on CT scans are intimately linked to the hemoglobin content within the hematoma [31]. In the initial phase of hemorrhage, red blood cells remain intact, resulting in a relatively low concentration of hemoglobin within the hematoma. Consequently, this manifests as low attenuation on CT images. Subsequently, as the red blood cells undergo lysis, hemoglobin is released, and the clot undergoes shrinkage, leading to the separation of serum [32]. This process significantly elevates the hemoglobin concentration, ultimately resulting in a high attenuation signal on CT. Consequently, a positive blend sign suggests the presence of a mobile liquid hematoma, which is more amenable to drainage, thereby facilitating HE.
Analysis of preoperative hematoma volume indicates that a larger hematoma is disadvantageous for HE within 24 hours postoperatively, with a threshold value of 45.46ml. The traditional surgical indication for cerebral hemorrhage is a hematoma volume of ≥ 30ml, and the standard surgical approach involves craniotomy for hematoma evacuation, optionally with decompressive craniectomy [18,19]. The Surgical Treatment for Intracerebral Hemorrhage (STICH) study suggested that craniotomy does not significantly improve the prognosis of patients [27,33]. Consequently, minimally invasive surgical techniques have emerged as viable alternatives. Nevertheless, given their inherent limitations, in cases involving substantial hematoma volumes, direct visualization of hematoma removal, such as through endoscopic or microscopic approaches, may still be necessary to enhance HE rates.
In addition, for sICH patients, more accurate catheterization may be needed. MISTIE II taught us that efficacy of the surgical task is directly related to satisfactory catheter placement [34]. Our profound research has also revealed that the precise placement of the catheter can effectively facilitate the early elimination of intracranial hematoma. Catheters that are accurately positioned within the hematoma enables surgeons to effectively aspirate the blood and clot, reducing the size of the hematoma and relieving pressure on surrounding tissues. Furthermore, as an NCCT marker, the irregular shape of hematoma is associated with hematoma expansion [24]. This may reflect the multi-source bleeding in the hematoma, along the multi-path development. The irregular shape of the hematoma can hinder the uniform distribution of urokinase and efficient drainage, ultimately compromising HE. Remarkably, our meticulous investigation has uncovered that the existence of a connection between the hematoma edge and the ventricle unexpectedly impedes HE, thus challenging the established clinical understanding. We offer a plausible hypothesis to explain this observation: the infiltration of cerebrospinal fluid into the hematoma cavity potentially dilutes the concentration of urokinase, subsequently limiting the dissolution of blood clots and ultimately compromising the efficiency of HE. In clinical practice, urokinase is often used to dissolve blood clots and transform solid hematoma into liquid hematoma. Studies have shown that urokinase can be used as a safe and effective alternative to alteplase [35], but the optimal dose of urokinase remains uncertain. In this study, all patients received urokinase in combination, but the effect of stereotactic aspiration combined with urokinase thrombolysis on long-term functional prognosis and the optimal dose of urokinase needed further study.
Drawing upon the aforementioned conclusions, we have developed an innovative predictive model. Within the framework of the early HE rates prediction model, patients scoring of 6 or 7 demonstrate a significantly elevated likelihood of achieving HE rates of 70% or higher. Similarly, within the context of the prediction model for HE rates prior to catheter removal, individuals scoring 2 or 3 exhibit superior HE outcomes.
Our findings are similar to the results of Kim JH et al. that a positive blend sign can promote the removal of hematoma in minimally invasive hematoma aspiration and catheter drainage [36]. This study was a single-center retrospective study, and our results further support this conclusion. In previous studies [36,37], the volume of the hematoma was calculated using the traditional formula [ABC/2], where A is the maximum diameter of the hematoma on the axial CT image, B is the diameter at 90° from A, and C is the number of CT sections in the hematoma multiplied by the section thickness. Although this method is simple and feasible, its accuracy is inadequate. Moreover, the CT value is measured only at the largest level of hematoma, which cannot reflect the average CT value of the whole hematoma. We have made modifications to our measurement methods. The accuracy of 3Dslicer in calculating the volume of hematoma is better than that of [ABC/2] [38], and 3Dslicer can automatically calculate the average CT value of the entire hematoma, indirectly reflecting the density of the entire hematoma. Wang T et al. believed that the CT value reflected the viscosity of the hematoma, the lower CT value was conducive to the removal of hematoma35. However, our results are the same as those of Kim JH et al. that hematoma CT values do not affect the HE. The reason may be different from the parameters set by CT instruments in different centers, resulting in bias in the results. In the future, unified CT parameters should be used to further determine the role of image features in predicting the removal of hematoma.
There are still some potential limitations in our study. Although it comes from a prospective, multicenter cerebral hemorrhage registry database, the sample size is small, which may affect the conclusion of the study and its universality. In addition, the predictive model employed in this study has yet to undergo external validation, necessitating future endeavors to thoroughly assess its performance on independent datasets.