A total of 125 consecutive SCH patients were screened in 4 neurosurgery departments from July 2019 to Nov. 2021. Among them, 38 cases failed to meet the inclusion criteria. Seven patients were lost to follow-up. Finally, 80 patients (36 MIC evacuation cases and 44 endoscopic evacuation cases) were included in the final analysis. A total of 349 CT scans were obtained and brought into analysis: 80 preoperative CT scans, 79 1-day postoperative scans, 79 3-day postoperative scans, 77 7-day postoperative scans, and 34 14-day postoperative scans. The ICCs values were 0.996 (95% CI: 0.995–0.997) for hematoma volume, and 0.972 (95% CI: 0.965–0.977) for PHE volume.
The mean (SD) age of the patients was 63.52 (10.2), with 44 males and 36 females. The mean (SD) interval time between onset and operation was 16.2 (5.9) hours. On admission, the median (IQR) GCS score was 11 (8–13). The mean (SD) preoperative hemorrhage volume was 21 (6.8), and the mean (SD) preoperative PHE volume was 10 (3.7). Among these patients, 69 (86.2%) had hypertension, 6 (7.5%) had diabetes, 10 (12.5%) had a history of craniocerebral disease, and 13 (16.2%) were smokers. Eleven serious adverse events (2 reoperation cases, 3 rebleeding cases, 2 intracranial infection cases, and 4 pulmonary infection cases) were observed during the follow-up.
Patients’ Prognoses
In the entire cohort, 61 patients (76.2%) had good neurological outcomes (mRS ≤ 3), while 19 patients (23.8%) had poor neurological outcomes (mRS ≥ 4). Demographic and clinical data for the patients with different neurological outcomes are shown in Table 1. Older patients (61.52 ± 10.58 vs. 69.89 ± 5.14 years, P = 0.001), with larger preoperative PHE volume (8.87 ± 3.23 vs. 13.52 ± 2.83 ml, P < 0.001), underwent MIC evacuation (62.30% vs. 31.58%, P = 0.037) were prone to had poor neurological outcome. The distributions of 6-month mRS scores in the current cohort are described in Fig. 1.
Risk Factors for Poor Neurological Outcome
Univariate logistic regression showed that older patients (P = 0.004), underwent MIC evacuation (P = 0.023), preoperative PHE volume (P < 0.001), 1-d postoperative PHE volume (P < 0.001), and 7-d postoperative PHE volume (P = 0.023) were correlated with poor neurological outcome (Table 2). These factors, together with 3-d postoperative PHE volume, were then entered into the multivariate logistic regression models to identify the independent predictors for poor neurological outcome.
Table 2
The correlation between the clinical factors and the 6-month prognosis.
Characteristic | | mRS Score ≥ 4 |
| Odds Ratio (95%CI) | P-value |
Age | | 1.11 (1.03–1.19) | 0.004 |
Pre-operative GCS score | | 1.01 (0.82–1.25) | 0.937 |
Interval Between Onset and Operation (hours) | | 0.96 (0.88–1.05) | 0.332 |
Surgical Methods (Reference is MIC Evacuation) | | 0.28 (0.09–0.84) | 0.023 |
Pre-operative Hematoma Volume (ml) | | 1.03 (0.95–1.11) | 0.484 |
Pre-operative PHE Volume (ml) | | 1.53 (1.25–1.88) | < 0.001 |
1d Post- operative Hematoma Volume (ml) | | 1.05 (0.942–1.16) | 0.400 |
1d Post- operative PHE Volume (ml) | | 1.42 (1.2–1.68) | < 0.001 |
3d Post-operative Hematoma Volume (ml) | | 0.85 (0.59–1.22) | 0.388 |
3d Post-operative PHE Volume (ml) | | 1.10 (0.98–1.22) | 0.109 |
7d Post-operative Hematoma Volume (ml) | | 0.97 (0.63–1.50) | 0.885 |
7d Post-operative PHE Volume (ml) | | 1.13 (1.02–1.25) | 0.023 |
14d Post-operative Hematoma Volume (ml) | | – | 0.362 |
14d Post-operative PHE Volume (ml) | | 0.96 (0.82–1.11) | 0.548 |
mRS: modified Rankin Scale. GCS: Glasgow Coma Scale. MIC: minimally invasive catheter. PHE: peri-hematoma edema. –: Not Applicable. |
Multivariate logistic regression showed that older patients, underwent MIC evacuation, large preoperative PHE, and large postoperative PHE were independent risk factors for poor neurological outcome (Fig. 2). Then, to explore the significance of preoperative PHE as a meaningful predictor of poor neurological outcome, ROC curves were plotted. The area under the ROC curve (AUC) of preoperative PHE was 0.849 when indicating the 6-month poor neurological outcome (Fig. 3). In the RCS curve determined to observe the relationship between preoperative PHE and prognosis, the risk of poor neurological outcome increased when preoperative PHE ≥ 10 ml (Fig. 4). Meanwhile, the Sankey diagram was used to described the relationship among preoperative GCS score, preoperative PHE, and 6-month mRS score (Fig. 5).
Thus, patients were divided into two subgroups: the large preoperative PHE subgroup (≥ 10 ml, n = 38) and the small preoperative PHE subgroup (< 10 ml, n = 42). In the small preoperative PHE subgroup, 41 patients (97.6%) had good neurological outcome, while 1 patient (2.4%) had poor neurological outcome. However, in the large preoperative PHE subgroup, 20 patients (52.6%) had good neurological outcome, while 18 patients (47.4%) had poor neurological outcome.
Evolution of hematoma and PHE after evacuation
In the entire cohort, the mean (SD) hemorrhage volume at postoperative day 1, 3, 7, and 14 were 4.76 (4.82) ml, 1.87 (1.78) ml, 0.93 (1.23) ml and 0.04 (0.12) ml, respectively. Among them, in MIC evacuation group, the mean (SD) hemorrhage volume at postoperative day 1, 3, 7, and 14 were 7.91 (4.96) ml, 2.03 (1.38) ml, 1.29 (1.33) ml, and 0.01 (0.01) ml, respectively. And in the endoscopic group, the mean (SD) hemorrhage volume at postoperative day 1, 3, 7, and 14 were 2.12 (2.65) ml, 1.73 (2.06) ml, 0.61 (1.06) ml and 0.06 (0.14) ml, respectively (Fig. 6A).
On the other hand, the mean (SD) PHE volume at postoperative day 1, 3, 7, and 14 were 11.45 (5.05) ml, 11.42 (4.77) ml, 11.98 (5.21) ml, and 10.84 (5.66) ml, respectively. Among them, in MIC evacuation group, the mean (SD) PHE volume at postoperative day 1, 3, 7, and 14 were 11.84 (6.08) ml, 10.26 (4.32) ml, 10.15 (4.19) ml, and 7.92 (2.55) ml, respectively. And in the endoscopic group, the mean (SD) PHE volume at postoperative day 1, 3, 7, and 14 were 11.28 (4.05) ml, 12.39 (4.97) ml, 13.60 (5.53) ml, and 12.43 (6.28) ml, respectively. Notably, from postoperative day 3, the PHE volume in the MIC evacuation group was smaller than that in the endoscopic evacuation group (Fig. 6B).