Effectiveness and Safety of Minimally Invasive Surgery for Infratentorial Hemorrhages: A Meta-Analysis of Clinical and Technical Outcomes

Abstract

Spontaneous infratentorial hemorrhages (SIHs) are a type of hemorrhagic stroke that compromise the cerebellum and/or brainstem. While minimally invasive surgical (MIS) approaches are effective in treating spontaneous supratentorial hemorrhage, their efficacy in managing SIH remains unclear. This uncertainty prompted the undertaking of this systematic review and meta-analysis. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a comprehensive search of the PubMed, Scopus, and Cochrane Library databases. Studies were included if they had a minimum of five patients and reported at least one outcome of interest. The outcomes evaluated were technical success, mean hematoma reduction, good functional outcomes at 3- and 6-month follow-up, complication rates, rebleeding rates, and postoperative mortality. A subgroup analysis was performed based on hemorrhage location and surgical approach. Depending on the data, either a random-effects or common-effects model was applied, and heterogeneity was assessed using I ² statistics. Seventeen studies met inclusion criteria. Postoperative mortality was 14.34% (95% confidence interval [CI]: 8.94–19.74%). Rate of good functional outcomes were 46.82% (95% CI: 18.88–74.77%) and 50.95% (95% CI: 32.72–69.17%) at 3 and 6 months, respectively. The overall complication rate was 6.86% (95% CI: 3.03–10.69%). Mortality by surgical technique ranged from 6.72% (miniature craniotomy only) to 26.27% (miniature craniotomy and catheter/puncture drainage), with no significant differences between approaches (p = 0.08). Patients with cerebellar hemorrhage experienced significantly greater rates of good functional outcome at 6 months compared with those with brainstem hemorrhage (p = 0.0004). Our study suggests that MIS for the treatment of SIH is an effective and feasible approach, as evidenced by low mortality rates and a limited number of associated complications. Larger randomized studies are warranted to compare different strategies and further optimize patient care.

Keywords

cerebellar - brainstem - MIS - hematoma - posterior fossa

Introduction

Spontaneous infratentorial hemorrhage (SIH) refers to the presence of a nontraumatic hematoma in the cerebellum or brainstem.[1] This type of bleeding accounts for 10 to 20% of all spontaneous intracerebral hemorrhages (sICHs) and can present with severe neurological deficits.[2] [3] sICH is associated with significant health care costs and mortality.[4] [5] [6]

Although SIH is less common than supratentorial hemorrhage, it is associated with a poorer prognosis and worse clinical outcomes.[7] [8] The reported mortality rate for cerebellar hemorrhages ranges from 18 to 75%, whereas brainstem hemorrhages exhibit a mortality rate between 30 and 90%.[7] [8]

Disability outcomes for survivors of SIH can be profound.[9] One cohort study demonstrated that patients with brainstem hemorrhages had a median modified Rankin Scale (mRS) score of 5 at discharge and 6 after 3 months of presenting with spontaneous cerebellar hematoma (SCH).[10] These findings are consistent with those reported in the INTERACT clinical trials, where an mRS score of 6 was more prevalent in the brainstem group compared with the cerebellum group.[9] [10] [11] [12] On the other hand, Witsch et al reported, in a systematic review, that the disability associated with SCH was moderate in 30% and severe in 33% of the patients, even in those who underwent surgical management.[13]

Current guidelines recommend surgical management for cerebellar hemorrhages through large unilateral or bilateral suboccipital craniectomy or craniotomy, with or without duraplasty and cerebrospinal fluid (CSF) diversion.[14] Although the medical literature on infratentorial hemorrhage evaluation is limited, studies suggest that surgical management is superior to medical treatment in patients with brainstem compression from cerebellar infarcts, as surgery is often associated with shorter recovery times.[15] [16]

Minimally invasive surgical (MIS) procedures, including endoscopic surgery, stereotactic aspiration, and miniature craniotomy, have demonstrated potential benefits in the management of spontaneous supratentorial hemorrhages.[17] These advantages include reduced surgical trauma, shorter recovery times, better neurological recovery, and decreased secondary complications.[17] [18] [19]

Despite significant evidence that MIS procedures are effective—and in some cases superior to traditional open surgeries, such as standard craniectomies or large craniotomies—for evacuating hematomas in supratentorial spontaneous hemorrhages, there is a lack of studies on the benefits of minimally invasive techniques for infratentorial hemorrhages, particularly SIH.[14] The objective of this study was to review the available literature and perform a pooled analysis of the effectiveness, safety, and clinical outcomes of MIS for SIH.

Material and Methods

This systematic review and meta-analysis adhered to the methodological standards outlined in the Cochrane Handbook for Systematic Reviews of Interventions and complied with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for its reporting structure. This study was registered with PROSPERO (https://www.crd.york.ac.uk/prospero/), registration number CRD42024568843.

Search Strategy and Data Extraction

We systematically searched PubMed, Scopus, and Cochrane Library from inception until July 2024, using the following search strategy (minimally invasive surgical procedures [Medical Subject Headings [MeSH]] OR “minimally invasive” OR minimally invasive OR “minimal surgical” OR endoscopy OR stereotaxy OR puncture OR craniopuncture OR “surgical evacuation” OR “key hole craniotomy” OR “key hole” OR keyhole) AND (Cranial Fossa, Posterior[MeSH] OR infratentorial OR cerebellar OR “brain stem” OR brainstem) AND (hemorrhage OR hematoma OR bleed OR bleeding OR “intracranial hemorrhage”), the syntax was accommodated to the respective database.

Three authors (M.F., J.E.S., and M.G.Z.) independently and blindly screened for eligible articles and extracted the data following the predefined search criteria. Missing data was excluded from final statistical analysis. Conflicts were then resolved through discussion among the authors.

Eligibility Criteria

The inclusion criteria were as follows: Randomized and nonrandomized studies reporting on SIHs treated with a MIS strategy were included. MIS was defined as robotic drainage, endoscopic drainage, the use of neuronavigation-assisted devices, or stereotactic procedures. Additionally, minimal cranial openings were defined as keyhole craniotomies less than 3 cm or puncture drainage techniques. Studies were included if at least one of the following were reported after MIS surgery: functional outcome, postoperative mortality, surgical complications, rebleeding after the procedure, postoperative technical success, and mean hematoma reduction. Studies that reported at least one relevant outcome were included and categorized accordingly in the respective analyses. Studies with less than five patients as well as case reports, animal studies, reviews, meta-analyses, systematic reviews, letters to the editor, abstracts, and non-English publications were excluded.

Outcome Measures

Good functional outcome was defined as a mRS score of less than or equal to 3 or a Glasgow Outcome Scale (GOS) greater than or equal to 4.[20] [21] The functional outcomes were analyzed at 3 and 6 months postintervention. Procedure-related complications were defined as any complications or adverse events that took place intraoperatively or related to the procedure (i.e., meningitis, cerebral abscess, the necessity for conversion to craniotomy or craniectomy, reintervention, CSF leak, cerebral lesions, rebleeding, etc.).[22] [23] Finally, technical success was defined as postoperative hematoma volume evacuation of 90% or greater, as evidence suggests that evacuation rates above or equal to this threshold are associated with improved functional outcomes.[24] Furthermore, the mean hematoma evacuation volume was also considered.

Quality Assessment

Nonrandomized studies were appraised with the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool.[25] The ROB2 Cochrane tool was used for randomized studies.[26]

Statistical Analysis

Given the noncomparative nature of the data set, a standard meta-analysis was not statistically feasible. An alternative to this method is a meta-analysis of proportions, where the weighted proportion of a binary outcome variable is formed by the average proportions of studies weighted by the inverse sampling variance. All analysis was performed in RStudio (RStudio Team [2024]; RStudio: Integrated Development Environment for R, RStudio, PBC, Boston, Massachusetts; URL: http://www.rstudio.com/) using the metaphor, weightr and meta packages.

A random-effects model was used to analyze the results. Results were presented as percentages with a 95% confidence interval (CI). High heterogeneity was defined as I ² > 50%. Final forest plots with summary subgroup effect sizes were then created using the meta package. Significant differences between subgroups were defined as a p-value of less than 0.05.

Results

Study Selection

A total of 2,190 articles were identified: 1,268 from PubMed, 912 from Scopus, and 10 from the Cochrane Library clinical trial registry. After removing 112 duplicates, 2,078 unique articles remained. Of these, 2,059 were excluded based on title or abstract, leaving 19 articles selected for full-text review (see [Fig. 1]). Ultimately, 17 studies were included in the final analysis, encompassing a total of 568 patients.

Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of study screening and selection.

Baseline Characteristics of Included Studies

As detailed in [Table 1], we included 17 studies comprising a total of 568 patients,[16] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] of which 306 (54%) were male. Thirteen of the studies were retrospective cohort studies or case series.[16] [27] [28] [29] [30] [31] [32] [33] [36] [37] [39] [41] [42] Three prospective nonrandomized studies[34] [35] [40] and one randomized study were included.[38] Thirteen studies included patients with cerebellar hemorrhage[16] [27] [28] [29] [30] [31] [33] [34] [37] [38] [39] [40] [41] and 4 studies evaluated patients with brainstem hemorrhage.[32] [35] [36] [42] The mean age of the patients ranged from 48.98 to 73 years. The mean hematoma volume on presentation ranged from 7.6 to 31.2 mL. Mean presentation Glasgow Coma Scale (GCS) score ranged from 3 to 13. The surgical techniques used included endoscopic drainage, miniature craniotomy alone, miniature craniotomy with puncture/catheter drainage, and robot-assisted procedures. All patient hemorrhages were spontaneous/hypertensive in origin as the study's authors excluded via angiography the presence of aneurysms, arteriovenous malformations, tumors, or other underlying pathology.[16] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [39] [40] [41] [42] Follow-up duration ranged from 0 to 90 months.

Table 1 Demographic and operative characteristics of included studies

Study	Design	Number of patients[d]	M/F	Mean age in years (range/SD)[a]	Hematoma location	Surgical method	Mean hematoma volume on presentation, mL (range/SD)[a]	Mean GCS on presentation (range/SD)[a]	Follow-up, months (range)
Khattar et al, 2019[28]	R[b]	6	4/2	60.2 (44–69)	Cerebellar	Endoscopic	22.3 (15–30)	10 (4–13)	2.2[f] (1–3)
Atsumi et al, 2019[37]	R[b]	15	5/10	73 (61–86)	Cerebellar	Endoscopic	21.8 (14–45)	10 (4–14)	Discharge
Chen et al, 2021[36]	R[b]	52	33/19	50.2 (33–72)	Brainstem	Miniature craniotomy only	8.5 (3.6–16.9)	6 ± 2.1	N/R (3–52)
Kellner et al, 2019[27]	R[b]	10	3/7	64.1 (47–82)	Cerebellar	Miniature craniotomy only	25.4 ± 9.2	9 (3–15)	33[f] (3–90)
Jin and Yang, 2024 A[41]	R[b]	77	36/41	63.02 ± 11.56	Cerebellar	Robot-assisted drainage	14.34 ± 4.48	13 ± 2.09	NR
Jin and Yang, 2024 B[41]	R[b]	61	22/39	62.77 ± 7.51	Cerebellar	Miniature craniotomy only	15.46 ± 4.61	12.10 ± 2.26	NR
Wang et al, 2019[34]	P-NR[c]	21	16/5	66.7 (47–89)	Cerebellar	Miniature craniotomy with puncture/catheter drainage	18.5 ± 5.0	10 ± 3.3	6
Li et al, 2023[29]	R[b]	35	20/15	63.2 ± 6.0	Cerebellar	Endoscopic	16.4 ± 4	NR	6
Li et al, 2018 A[30]	R[b]	25	16/9	61.8 ± 6.7	Cerebellar	Endoscopic	15.9 ± 3.5	11.0 ± 3.0	1–12
Li et al,2018 B[30]	R[b]	23	13/10	62.5 ± 7.3	Cerebellar	Miniature craniotomy with puncture/catheter drainage	15.5 ± 3.6	12 ± 2.9	1–12
Tokimura et al, 2010[39]	R[b]	23	18/5	67.2 ± 2.02	Cerebellar	Miniature craniotomy only	30.2 ± 3.0	NR	Discharge
Li et al,2021 A[31]	R[b]	27	16/11	66 (40–80)	Cerebellar	Endoscopic	21.91 ± 7.21	10 ± 2.56	6
Li et al,2021 B[31]	R[b]	25	14/11	68 (43–63)	Cerebellar	Miniature craniotomy with puncture/catheter drainage	21.56 ± 8.64	9.00 ± 3.07	6
Tang et al, 2024[42]	R[b]	51	35/16	55 (49–63)	Brainstem	Robot-assisted drainage	7.6 (6.5–9.7)	4 (3–5)	6
Wang et al, 2022[35]	P-NR[c]	5	4/1	57 (34–70)	Brainstem	Miniature craniotomy with puncture/catheter drainage	10 (7–18)	5 (3–7)	NR
Yamamoto et al, 2024[40]	P-NR[c]	20	N/R	N/R	Cerebellar	Endoscopic	24.2 ± 12.3	NR	NR
Bao et al, 2024[32]	R[b]	36	24/12	48.98 ± 9.89	Brainstem	Robot-assisted drainage	10.8 ± 5.31	3 (3–5)	2
Yoh et al, 2023[33]	R[b]	5	2/3	65 ± 56–81	Cerebellar	Miniature craniotomy only	16 (12–20)	12 (8–15)	31[f] (1–66)
Tamaki et al, 2004[38]	P-R[e]	25	18/7	63 ± 14	Cerebellar	Miniature craniotomy only	31.2 ± 16.7	NR	3
Lee et al, 2012[16]	R[b]	26	7/19	64.7 ± 9.4	Cerebellar	Miniature craniotomy with puncture/catheter drainage	21.8 ± 5.8	13 ± 1.3	23.7[f]

Abbreviations: F, female; GCS, Glasgow Coma Scale; M, male; N/R, not reported; SD, standard deviation.

Note: Studies that are listed twice with different data sets are done so for those with different surgical techniques to avoid confusion in the subgroup analysis.

^a Numbers accompanied by a range are medians and those with a ± symbol are means.

^b Retrospective.

^c Prospective nonrandomized.

^d Number of patients included in analysis.

^e Prospective randomized.

^f Mean months of follow-up.

Technical Outcomes: Mean Hematoma Reduction and Technical Success

The mean hematoma reduction percentage was reported by nine studies; however, only four could be included in the proportional analysis of technical success due to completeness of data.[27] [28] [29] [30] [31] [33] [37] [38] [39] The overall technical success rate of MIS techniques for SIH was 71.36% (95% CI: 48.64–94.08%; I ²= 70%; [Fig. 2]). The mean reduction in hematoma volume was 11.78 mL (95% CI: 6.17–17.40%; I ²= 92.2%).

Fig. 2 Forest plot for technical success.

Clinical Outcomes: Functional Outcome, Complications, and Overall Mortality

A good functional outcome rate in the first 3 months postsurgery was reported by seven studies,[28] [32] [33] [36] [37] [38] [39] which was found to be 46.82% (95% CI: 18.88–74.77%; I ²= 97%; [Fig. 3A]) and at 6 months postoperation was 50.95% (95% CI: 32.72–69.17%; I ²= 94%; [Fig. 3B]) as reported by six studies.[16] [29] [31] [34] [40] [42] The rate of postoperative mortality was 14.34% (95% CI: 8.94–19.74%; I ²= 67%; [Fig. 4A]). The rate of procedure-related complication was 6.86% (95% CI: 3.03–10.69%; I ²= 66%; [Fig. 4B]).

Fig. 3 (A) Forest plot for good functional outcome in the first 3 months of follow-up. (B) Forest plot for good functional outcome at 6 months of follow-up.

Fig. 4 (A) Forest plot for postoperative mortality. (B) Forest plot for postintervention complication rate.

Subgroup Analysis by Location and Surgical Technique

A subgroup analysis was conducted ([Fig. 5]) to evaluate overall mortality rates across different MIS techniques. The overall mortality rate for the endoscopic surgery group was 13.35% (95% CI: 7.01–19.70%; I ² = 0%). For the miniature craniotomy-only group, the rate was 6.72% (95% CI: 2.60–10.85%; I ² = 0%). The robotic-drainage group had a mortality rate of 12.37% (95% CI: 1.83–22.91%; I ² = 79%). The miniature craniotomy with puncture/catheter drainage group showed a mortality rate of 26.27% (95% CI: 7.71–44.84%; I ² = 86%). Subgroup comparisons revealed no statistically significant differences between the subgroups (p = 0.08). Furthermore, the complication rate was also analyzed by surgical technique ([Fig. 6]). The pooled complication rate for the endoscopic surgery group was 11.07% (95% CI: 2.91–19.23%; I ² = 60.4%). For the miniature craniotomy-only group, the rate was 5.53% (95% CI: 0.00–13.13%; I ² = 71.5%). The robotic-drainage group had a complication rate of 1.93% (95% CI: 0.00–6.25%; I ² = 65.3%). The miniature craniotomy with puncture/catheter drainage group showed a complication rate of 10.37% (95% CI: 0.42–20.33%; I ² = 67.9%). Also, subgroup comparisons revealed no statistically significant differences between the subgroups (p = 0.15). Finally, the surgical technique that demonstrated the greatest reduction in hematoma volume was miniature craniotomy only, with a mean reduction of 22.6 mL (95% CI: 16.45–28.75) ([Fig. 7]), which was significantly greater, compared with miniature craniotomy with puncture/catheter drainage with a mean reduction of 8.50 mL (95% CI: 2.82–14.18, I ² = 88.1%; p = 0.0005).

Fig. 5 Forest plot for subgroup analysis of postoperative mortality by surgical technique.

Fig. 6 Forest plot for subgroup analysis of postoperative complications by surgical technique.

Fig. 7 Forest plot for subgroup analysis of mean hematoma volume reduction by surgical technique.

Postoperative mortality rates were analyzed based on hemorrhage location ([Fig. 8]). For cerebellar hemorrhages, the rate was 14.36% (95% CI: 7.26–21.46%; I ² = 71%). For brainstem hemorrhages, the rate was 16.34% (95% CI: 10.32–22.37%; I ² = 0%). The subgroup comparison did not reveal a statistically significant difference between the two groups (p = 0.676). In terms of good function outcomes, there was no statistically significant difference between those with cerebellar or brainstem hemorrhages at 3 months' follow-up, with rates of 59.33% (95% CI: 26.98–91.68%; I ² = 95%) and 17.71% (95% CI: 0.00–53.50%; I ² = 97.2%) for cerebellar and brain stem hemorrhages, respectively ([Fig. 9A]). However, there was a statistically significant result in favor of cerebellar hemorrhages at 6-month follow-up with a good functional outcome of 56.93% (95% CI: 39.50–74.35%; I ² = 89%) compared with 19.61% (95% CI: 9.82–33.12%) for brainstem hemorrhages, p = 0.0004 ([Fig. 9B]).

Fig. 8 Forest plot for subgroup analysis of postoperative mortality by hemorrhage location.

Fig. 9 (A) Forest plot for subgroup analysis of good functional outcome at 3 months by hemorrhage location. (B) Forest plot for subgroup analysis of good functional outcome at 6 months by hemorrhage location.

Risk of Bias Assessment

Of the 16 nonrandomized studies included, 4 studies received an overall ROBINS-I rating of “moderate” risk of bias and 11 studies received “serious” risk of bias, mainly due to the absence of control groups and the potential for unmeasured confounders. The randomized study received an overall risk of bias rating of ̈high.” The individual appraisal of studies is reported in [Supplementary Figs. S1] and [S2], available in the online version.

Discussion

This is the first single-arm meta-analysis evaluating the technical and clinical outcomes of MIS techniques for managing infratentorial sICH. Our results indicate that MIS is a valuable procedure with acceptable complication rates for a condition associated with high complications and mortality.

Although we included both cerebellar and brainstem locations under the term SIH, these entities do not necessarily share the same surgical indications and operative strategies. Cerebellar hematomas are typically approached to relieve posterior fossa mass effect and brainstem compression. In contrast, primary brainstem hemorrhages are inherently more complex, have a poorer prognosis, and are usually considered for surgery only in highly selected cases, with a focus on focal evacuation rather than decompression. Therefore, the same operative goals or indications do not apply uniformly across both locations. In this meta-analysis, the decision to analyze them within a common framework was made to address the overarching question of whether minimally invasive approaches can be performed safely and effectively in the infratentorial compartment at all, while still emphasizing that decision-making must remain location-specific.

Technical and Clinical Outcomes

The latest American Heart Association guidelines recommend urgent surgical hematoma evacuation for patients with hemorrhage of the posterior fossa region; however, no specific recommendation is made regarding MIS over traditional surgery.[14] Infratentorial hemorrhages are associated with an inherently poor prognosis, as evidenced by various cohorts managed with traditional, nonminimally invasive procedures such as suboccipital craniectomy, where postoperative mortality rates can exceed 50% for brainstem hemorrhages.[43] [44] [45] In contrast, our review found a significantly lower mortality rate of 14.34% (95% CI: 8.94–19.74%; I ²= 67%; [Fig. 4A]) suggesting that MIS may offer a more viable option for these patients. It is important to note, however, that the studies included in this review did not specify exactly at what point during follow-up postoperative death occurred, and some patients had baseline pathologies and medical complications not related to the surgical technique such as pulmonary infections that further complicated their prognosis.[16] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] Moreover, two studies could not be included in the mortality analysis as one study which included supratentorial and infratentorial hemorrhage did not specify the location of hemorrhage of the deceased patients.[40] Another study grouped patients with a postoperative GOS score of 1 to 3 but did not specify the exact number of fatalities.[39]

Determining the benefits of MIS for improving functional outcomes remains more challenging. In this study, it was observed that minimally invasive surgery achieved acceptable rates of functional outcomes, with 46.82% (95% CI: 18.88–74.77%; I ²= 97%; [Fig. 3]) at 3 months and 50.95% (95% CI: 32.72–69.17%; I ²= 94%; [Fig. 4]) at 6 months postintervention. These findings were compared with a systematic review and meta-analysis evaluating surgical versus conservative management for cerebellar hemorrhage.[46] Kuramatsu et al reported a 30.9% rate of good functional outcomes, defined as a mRS score of 0 to 3, at 3 months; their analysis included traditional and MIS techniques without performing a subgroup analysis.[46] However, another systematic review evaluating clinical outcomes in patients with cerebellar hemorrhage managed with various approaches—including unspecified surgical drainage techniques and medical management—reported a good functional outcome rate similar to ours, at 42% at 6 months.[47]

Low rates of procedure-related complications were observed in our study, indicating that MIS is a safe procedure and a good option in infratentorial hemorrhages. In this meta-analysis, a total of 52 complications were reported, including intracranial infection, rebleeding, CSF leak, and the requirement of conversion to open craniotomy or reintervention. Among the complications associated with the procedure, intracranial infections and rebleeding were the most common, each accounting for 39 and 36% of all complications, respectively, followed by reintervention/conversion of procedure, which accounted for 13% of the complications. Finally, the least common complication was CSF leak, which made up the remaining 12%. Central fever was frequently noted, particularly in studies involving patients with brainstem hemorrhages; however, this is more related to the pathophysiology of the lesion itself.[48] Comparisons with complications reported in the literature revealed similar results. A multicenter randomized controlled trial evaluating stereotactic computed tomography-guided endoscopic surgery for supratentorial hemorrhages reported that 35% of their complications were due to rebleeding, and meningitis was less common, occurring in 15% of cases.[49]

Three of the included studies in this review compared an MIS technique to a conventional craniotomy/craniectomy technique and all studies found higher rates of meningitis, CSF leak, and/or rebleeding in the conventional craniotomy/craniectomy groups.[29] [37] [38] In addition, Atsumi et al found that three patients in the conventional craniotomy group required reintervention, compared with none in the MIS group.[37] These initial results suggest that MIS is a safer surgical approach compared with conventional craniotomy/craniectomy.

Subgroup Analysis

There were four different types of interventions performed in the studies included in our meta-analysis, which were miniature craniotomy only, miniature craniotomy and puncture/catheter drainage, endoscopy, and robotic drainage. The procedure associated with the lowest mortality was the miniature craniotomy-only group with an overall postoperative mortality rate of 6.72% (95% CI: 2.60–10.85%; I ² = 0%) and the highest mortality was found in the miniature craniotomy and puncture/catheter drainage group with a 26.27% (95% CI: 7.71–44.84%; I ² = 86%). Although the difference between the subgroups was not statistically significant, an absolute reduction of almost 20% between the use of catheter/puncture drainage and not using these adjuncts may be of clinical significance. It is important to note that there were no large differences between these two groups that may have contributed to a selection bias for sicker patients, accounting for the higher mortality. For example, the presenting GCS for the miniature craniotomy group ranged from 6 to 12, whereas for the miniature craniotomy and puncture/catheter drainage group, the range was 5 to 13.[16] [27] [30] [31] [33] [34] [35] [36] [38] [41] Furthermore, the mean hematoma volume ranged from 15.46 to 31.2 and 10 to 21.8 mL for the miniature craniotomy and the miniature craniotomy and puncture/catheter drainage group, respectively.[16] [27] [30] [31] [33] [34] [35] [36] [38] [41] Moreover, greater hematoma volume reduction found in the miniature craniotomy group may be a contributing factor to the magnitude of the differences observed in mortality.

Overall, MIS was associated with a low rate of procedure-related complications. The lowest complication rate was observed in the robotic drainage subgroup; however, this group included the fewest patients, leading to imprecise estimates and wide CIs. Despite this, robotic-assisted neurosurgical techniques have generally demonstrated low complication rates across multiple indications, and infratentorial hematoma evacuation may share a similar safety profile in carefully selected cases.[50]

In terms of hemorrhage location, we did not observe a statistically significant difference in postoperative mortality between cerebellar and brainstem hemorrhages; however, these estimates must be interpreted with caution. The cerebellar cohort comprised the majority of patients, whereas only a small number of cases of brainstem hemorrhage were reported across four studies. Under these conditions, similar point estimates of mortality do not imply that cerebellar and brainstem hemorrhages share comparable indications or risk–benefit profiles for surgery. Rather, our data suggest that, when MIS is chosen, perioperative mortality can be maintained within an acceptable range even in brainstem cases. Consistent with the inherently poorer prognosis of brainstem hemorrhages, we did find a significantly lower rate of good functional outcome at 6 months in this subgroup compared with cerebellar hematomas.[48] These findings reinforce that MIS for brainstem hemorrhage should still be considered an emerging, highly selective strategy, while MIS for cerebellar hemorrhage may apply to a broader population under established guideline-driven indications.

When it comes to the safety of the procedure, we found that all of the procedure-related adverse events reported were in the cerebellar hemorrhage group. It is important to note, however, that the cerebellar hemorrhage group had more than twice as many patients as the brainstem group, making these results difficult to interpret. However, more than 50% of these patients were treated with robotic drainage, indicating that this could be a safer option for patients with primary brainstem hemorrhage.

Limitations

Our study had several limitations. First, our risk of bias assessment revealed that most of the included studies had a serious risk of bias, primarily due to confounding variables. Furthermore, language bias may have contributed to the selection of included studies. Second, we were unable to analyze certain technical conditions, such as surgical timing, and patient conditions, such as obstructive hydrocephalus or ventricular compromise, due to significant heterogeneity in reporting. Third, the small size of the included cohorts reduced the power of the overall analysis. Additionally, the timing of surgery, which could impact postoperative mortality and disability, was not consistently reported among the studies, preventing analysis of this variable. Furthermore, some studies did not clearly state the presence or absence of adverse events, which excluded them from the relevant analysis. Moreover, although cerebellar and brainstem hemorrhages were analyzed together under the SIH construct, they represent biologically and surgically distinct entities with different indications and operative goals. We attempted to mitigate this by performing prespecified subgroup analyses by location, but the relatively small number of brainstem cases means that those results are hypothesis-generating and should not be extrapolated to broad clinical practice.

Future Directions

Future research should aim to overcome the methodological limitations of the current literature by prioritizing well-designed, prospective studies that directly compare minimally invasive techniques with conventional open surgery for infratentorial hemorrhages. Standardized reporting of critical clinical and technical parameters—such as hematoma volume, location, timing of intervention, degree of mass effect, and patient neurological status—will be essential to reduce heterogeneity and enable more meaningful subgroup analyses. Further efforts should also explore the influence of emerging technologies, including endoscopy, navigation, and robotics, to better delineate the specific technical advantages and limitations of MIS in the posterior fossa. By generating higher-quality evidence and consistent reporting standards, future studies can lay the groundwork for developing clear, evidence-based guidelines for the management of SIH.

Conclusion

MIS is a technically feasible, safe, and effective management strategy for SIH, especially in terms of mortality and procedure-related complications, particularly for cerebellar hemorrhages. Furthermore, although there were no statistically significant differences between the drainage techniques used, absolute differences in mortality are not negligible and warrant more studies that directly compare different drainage strategies. Larger randomized studies are required to enhance clinical relevance, focusing on criteria for selecting MIS over traditional, more invasive surgery (considering hematoma volume, location, patient age, comorbidities, clinical presentation, among others). These factors need further exploration to develop a synthesis suitable for clinical guidelines.

Conflict of Interest

None declared.

• References

• 1 Aguilar MI, Freeman WD. Spontaneous intracerebral hemorrhage. Semin Neurol 2010; 30 (05) 555-564
• 2 Ortiz de Mendivil A, Alcalá-Galiano A, Ochoa M, Salvador E, Millán JM. Brainstem stroke: anatomy, clinical and radiological findings. Semin Ultrasound CT MR 2013; 34 (02) 131-141
• 3 Fischer MA, Das JM. Cerebellar Hematoma. In: StatPearls. StatPearls Publishing; 2024. . Accessed August 19, 2024 at: http://www.ncbi.nlm.nih.gov/books/NBK541076/
• 4 Fernando SM, Reardon PM, Dowlatshahi D. et al. Outcomes and costs of patients admitted to the ICU due to spontaneous intracranial hemorrhage. Crit Care Med 2018; 46 (05) e395-e403
• 5 Smeds M, Skrifvars MB, Reinikainen M. et al. One-year healthcare costs of patients with spontaneous intracerebral hemorrhage treated in the intensive care unit. Eur Stroke J 2022; 7 (03) 267-279
• 6 Specogna AV, Patten SB, Turin TC, Hill MD. Cost of spontaneous intracerebral hemorrhage in Canada during 1 decade. Stroke 2014; 45 (01) 284-286
• 7 Pong V, Chan KH, Chong BH. et al. Long-term outcome and prognostic factors after spontaneous cerebellar hemorrhage. Cerebellum 2012; 11 (04) 939-945
• 8 Chen D, Tang Y, Nie H. et al. Primary brainstem hemorrhage: a review of prognostic factors and surgical management. Front Neurol 2021; 12: 727962
• 9 Woo D, Comeau ME, Venema SU. et al. Risk factors associated with mortality and neurologic disability after intracerebral hemorrhage in a racially and ethnically diverse cohort. JAMA Netw Open 2022; 5 (03) e221103
• 10 Hegde A, Menon G, Kumar V. et al. Clinical profile and predictors of outcome in spontaneous intracerebral hemorrhage from a tertiary care centre in South India. Stroke Res Treat 2020; 2020: 2192709
• 11 Chen R, Wang X, Anderson CS. et al. Infratentorial intracerebral hemorrhage. Stroke 2019; 50 (05) 1257-1259
• 12 Anderson CS, Huang Y, Wang JG. et al; INTERACT Investigators. Intensive blood pressure reduction in acute cerebral haemorrhage trial (INTERACT): a randomised pilot trial. Lancet Neurol 2008; 7 (05) 391-399
• 13 Witsch J, Neugebauer H, Zweckberger K, Jüttler E. Primary cerebellar haemorrhage: complications, treatment and outcome. Clin Neurol Neurosurg 2013; 115 (07) 863-869
• 14 Greenberg SM, Ziai WC, Cordonnier C. et al. 2022 guideline for the management of patients with spontaneous intracerebral hemorrhage: a guideline from the American Heart Association/American Stroke Association. Stroke 2022; 53 (07)
• 15 Mostofi K. Neurosurgical management of massive cerebellar infarct outcome in 53 patients. Surg Neurol Int 2013; 4 (01) 28
• 16 Lee JH, Kim DW, Kang SD. Stereotactic burr hole aspiration surgery for spontaneous hypertensive cerebellar hemorrhage. J Cerebrovasc Endovasc Neurosurg 2012; 14 (03) 170-174
• 17 Xia Z, Wu X, Li J. et al. Minimally invasive surgery is superior to conventional craniotomy in patients with spontaneous supratentorial intracerebral hemorrhage: a systematic review and meta-analysis. World Neurosurg 2018; 115: 266-273
• 18 Jianhua X, Zhenying H, Bingbing L. et al. Comparison of surgical outcomes and recovery of neurologic and linguistic functions in the dominant hemisphere after basal ganglia hematoma evacuation by craniotomy versus endoscopy. World Neurosurg 2019; 129: e494-e501
• 19 Hou D, Lu Y, Wu D, Tang Y, Dong Q. Minimally invasive surgery in patients with intracerebral hemorrhage: a meta-analysis of randomized controlled trials. Front Neurol 2022; 12: 789757
• 20 van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988; 19 (05) 604-607
• 21 McMillan T, Wilson L, Ponsford J, Levin H, Teasdale G, Bond M. The Glasgow Outcome Scale - 40 years of application and refinement. Nat Rev Neurol 2016; 12 (08) 477-485
• 22 Zhang G, Pan C, Zhang P. et al. Precision of minimally invasive surgery for intracerebral hemorrhage treatment. Brain Hemorrhages 2020; 1 (04) 200-204
• 23 Ibrahim A, Arifianto MR, Al FauziA. Minimally invasive neuroendoscopic surgery for spontaneous intracerebral hemorrhage: a review of the rationale and associated complications. In: Turel KE, Chernov MF, Sarkar H. eds. Complications in Neurosurgery. Vol 130. Acta Neurochirurgica Supplement. Vienna, Austria: Springer International Publishing; 2023: 103-108
• 24 Choy DKS, Wu PH, Tan D, Yeo TT, Chou N. Correlation of the long-term neurological outcomes with completeness of surgical evacuation in spontaneous supratentorial intracerebral haemorrhage: a retrospective study. Singapore Med J 2010; 51 (04) 320-325
• 25 Sterne JA, Hernán MA, Reeves BC. et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919
• 26 Sterne JAC, Savović J, Page MJ. et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898
• 27 Kellner CP, Moore F, Arginteanu MS. et al. Minimally invasive evacuation of spontaneous cerebellar intracerebral hemorrhage. World Neurosurg 2019; 122: e1-e9
• 28 Khattar NK, Fortuny EM, Wessell AP. et al. Minimally invasive surgery for spontaneous cerebellar hemorrhage: a multicenter study. World Neurosurg 2019; 129: e35-e39
• 29 Li Q, Duan F, Luo M, Luo Z. Neuronavigation-assisted neuroendoscopy versus conventional craniotomy for hypertensive cerebellar hemorrhage: a comparative analysis of efficacy and outcomes. Clin Neurol Neurosurg 2023; 233: 107960
• 30 Li L, Li Z, Li Y. et al. Surgical evacuation of spontaneous cerebellar hemorrhage: comparison of safety and efficacy of suboccipital craniotomy, stereotactic aspiration, and thrombolysis and endoscopic surgery. World Neurosurg 2018; 117: e90-e98
• 31 Li L, Liu H, Luo J. et al. Comparison of long-term outcomes of endoscopic and minimally invasive catheter evacuation for the treatment of spontaneous cerebellar hemorrhage. Transl Stroke Res 2021; 12 (01) 57-64
• 32 Bao D, Ni S, Chang B, Zhang W, Zhang H, Niu C. Short-term outcomes of robot-assisted minimally invasive surgery for brainstem hemorrhage: a case-control study. Heliyon 2024; 10 (04) e25912
• 33 Yoh N, Abou-Al-Shaar H, Bethamcharla R. et al. Minimally invasive surgical evacuation for spontaneous cerebellar hemorrhage: a case series and systematic review. Neurosurg Rev 2023; 46 (01) 208
• 34 Wang J, Wu QY, Du CP. et al. Spontaneous cerebellar hemorrhage with severe brainstem dysfunction through minimally invasive puncture treatment by locating the simple bedside. Medicine (Baltimore) 2019; 98 (38) e17211
• 35 Wang Q, Guo W, Zhang T. et al. Laser navigation combined with XperCT technology assisted puncture of brainstem hemorrhage. Front Neurol 2022; 13: 905477
• 36 Chen LH, Li FJ, Zhang HT, Chen WJ, Sun K, Xu RX. The microsurgical treatment for primary hypertensive brainstem hemorrhage: Experience with 52 patients. Asian J Surg 2021; 44 (01) 123-130
• 37 Atsumi H, Baba T, Sunaga A. et al. Neuroendoscopic evacuation for spontaneous cerebellar hemorrhage is a safe and secure approach and may become a mainstream technique. Neurol Med Chir (Tokyo) 2019; 59 (11) 423-429
• 38 Tamaki T, Kitamura T, Node Y, Teramoto A. Paramedian suboccipital mini-craniectomy for evacuation of spontaneous cerebellar hemorrhage. Neurol Med Chir (Tokyo) 2004; 44 (11) 578-582 , discussion 583
• 39 Tokimura H, Tajitsu K, Taniguchi A. et al. Efficacy and safety of key hole craniotomy for the evacuation of spontaneous cerebellar hemorrhage. Neurol Med Chir (Tokyo) 2010; 50 (05) 367-372
• 40 Yamamoto T, Watabe T, Yamashiro S. et al; RICH-Trend Investigators. Safety of endoscopic surgery for spontaneous intracerebral hemorrhage in the registry of intracerebral hemorrhage treated by endoscopic hematoma evacuation in Japan. World Neurosurg 2024; 189: e370-e379
• 41 Jin C, Yang Y. Surgical evacuation of spontaneous cerebellar hemorrhage: comparison of safety and efficacy of suboccipital craniotomy and robotic-assisted stereotactic hematoma drainage. Clin Neurol Neurosurg 2024; 239: 108192
• 42 Tang Z, Huang W, Chen Q. et al. Curative effect analysis of robot-assisted drainage surgery in treatment of spontaneous hypertensive brainstem hemorrhage. Front Neurol 2024; 15: 1352949
• 43 Da Pian R, Bazzan A, Pasqualin A. Surgical versus medical treatment of spontaneous posterior fossa haematomas: a cooperative study on 205 cases. Neurol Res 1984; 6 (03) 145-151
• 44 Sviri GE, Hayek S, Paldor I. Spontaneous cerebellar hemorrhage carries a grim prognosis in both operated and unoperated patients. J Clin Neurosci 2020; 78: 121-127
• 45 Hackenberg KAM, Unterberg AW, Jung CS, Bösel J, Schönenberger S, Zweckberger K. Does suboccipital decompression and evacuation of intraparenchymal hematoma improve neurological outcome in patients with spontaneous cerebellar hemorrhage?. Clin Neurol Neurosurg 2017; 155: 22-29
• 46 Kuramatsu JB, Biffi A, Gerner ST. et al. Association of surgical hematoma evacuation vs conservative treatment with functional outcome in patients with cerebellar intracerebral hemorrhage. JAMA 2019; 322 (14) 1392-1403
• 47 Singh SD, Brouwers HB, Senff JR. et al. Haematoma evacuation in cerebellar intracerebral haemorrhage: systematic review. J Neurol Neurosurg Psychiatry 2020; 91 (01) 82-87
• 48 Samudra N, Figueroa S. Intractable central hyperthermia in the setting of brainstem hemorrhage. Ther Hypothermia Temp Manag 2016; 6 (02) 98-101
• 49 Vespa P, Hanley D, Betz J. et al; ICES Investigators. ICES (intraoperative stereotactic computed tomography-guided endoscopic surgery) for brain hemorrhage: a multicenter randomized controlled trial. Stroke 2016; 47 (11) 2749-2755
• 50 Fomenko A, Serletis D. Robotic stereotaxy in cranial neurosurgery: a qualitative systematic review. Neurosurgery 2018; 83 (04) 642-650

Publication History

Article published online:
03 February 2026