Functional Outcomes After Decompressive Surgery in Patients with Malignant Space-Occupying Cerebellar Infarction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a retrospective study on  58 patients who underwent posterior fossa  decompressive surgery for acute cerebrllar infarctio.  The 27 90-day mortality rate was 27.6% . The analysis has been done classifying the patients in good (mRS 0-3)  and poor (mRS>3) outcome. Good functional outcome  was achieved in 24 patients (41.4%). Multivariable analysis showed, as factors associted to  poor outcome on day 90 (mRS 4-6): higher premorbid mRS score, higher NIHSS score on admission  and the presence of an additional brainstem infarction. moreover, hyperactive delirium was associated with good clinical outcome (OR 0.020 [95%CI, 0.001 – 0.623]; p=0.026).

The Authors conclude that decompressive surgery is a safe, life-saving treatment for malignant swelling in space-occupying cerebellar  infarction.

The paper is of practical interest. Apart from the analysis dividing subjects into two groups, an ordinal-shift analysis could provide more information. The association of hyperactive delirium and good outcome deserves deeper discussion 

Author Response

Thank you very much for taking the time to review this manuscript. The authors have considered the comments and tried our best to address them. We hope the manuscript after careful revision meet your standards.

The authors welcome further constructive comments if any.

Below we provide the point-by-point responses. All modifications in the manuscript have been highlighted in red.

Comment 1:

This is a retrospective study on 58 patients who underwent posterior fossa decompressive surgery for acute cerebellar infarction. The 90-day mortality rate was 27.6% . The analysis has been done classifying the patients in good (mRS 0-3) and poor (mRS>3) outcome. Good functional outcome was achieved in 24 patients (41.4%). Multivariable analysis showed, as factors associated to poor outcome on day 90 (mRS 4-6): higher premorbid mRS score, higher NIHSS score on admission and the presence of an additional brainstem infarction. moreover, hyperactive delirium was associated with good clinical outcome (OR 0.020 [95%CI, 0.001 – 0.623]; p=0.026).

The Authors conclude that decompressive surgery is a safe, life-saving treatment for malignant swelling in space-occupying cerebellar infarction.

The paper is of practical interest. Apart from the analysis dividing subjects into two groups, an ordinal-shift analysis could provide more information.

Response: We thank the reviewer for the positive comments on our manuscript. As for the suggestion concerning the statistical analysis, ordinal regression deals with dependent variables that have multiple levels. However, we chose to dichotomize the outcome variable mRS into "good outcome " (mRS 0-3) and "bad outcome " (mRS 4-6) instead of using all levels as we do not have an abundance of patients in the study. With a 7-level mRS, some levels will not contain enough patients for the model coefficients to converge. We are confident that the crude division into "good" and "bad" mRS shows enough information.

Comment 2:

The association of hyperactive delirium and good outcome deserves deeper discussion. 

Response :

We thank the reviewer for this valuable comment and appreciate the opportunity to clarify this issue that caused irritation among many readers of the manuscript.

Unexpectedly, the occurrence of postoperative hyperactive delirium was statistically associated with a good functional outcome. This contradicts the known negative impact of delirium on the prognosis of patients with acute stroke. Previous studies demonstrated that patients with post-stroke delirium have a worse functional outcome and higher mortality. We ascribe the result of our study to the effect of the so-called “survivorship bias”, an error that occurs when researchers focus on groups or cases that have passed a certain selection process. Patients in the “poor functional outcome group” were often in such a bad clinical state, i.e., dead within a few days or unconscious, that they were unable to express symptoms of a delirium. Thus, delirium remained an underrepresented complication in this group. Nevertheless, amid the small sample size of only 8 cases of reported hyperactive delirium, we emphasize that this result must be treated with caution and warrant verification in a larger prospective study.

In order to avoid further confusion or misinterpretation, we added this paragraph to the section Discussion (manuscript page 8, lines 182 - 195). 

 

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript titled "Functional outcome after decompressive surgery in patients with malignant space-occupying cerebellar infarction" investigates the outcomes of decompressive surgery in patients with this condition. The study retrospectively analyzes 58 patients who underwent surgery, focusing on factors influencing their recovery and complications during hospitalization. Key findings include a 27.6% 90-day mortality rate and that higher premorbid mRS, higher NIHSS scores on admission, and the presence of brainstem infarction were associated with poor outcomes (mRS 4-6). The study suggests that despite a significant portion of patients experiencing poor outcomes, decompressive surgery remains a crucial life-saving option.

 

1. The results of the multivariable logistic regression could benefit from a more detailed explanation in the results section. Specifically, further discussion on why hyperactive delirium was associated with better outcomes would strengthen the manuscript.

 

2. The manuscript used Kaplan-Meier analysis for evaluating survival outcomes. Accordingly, Cox proportional hazards models should be more appropriate than logistic regression models for analyzing survival data.

 

3. The discussion effectively interprets the findings but should provide a more in-depth future plan. For instance, the authors should add the potential to detect penumbral lesions using the advanced MRI technique. The following paper is useful for explaining this technology:

10.3389/fneur.2022.752450

 

4. Other MRI techniques can predict the clinical outcome in these patients. Longitudinal changes in iron and myelination within Ischemic lesions detected by QSM were associated with neurological outcomes. Please consider to add it to the Discussion:

 

10.1161/STROKEAHA.123.044606 

Author Response

Thank you very much for taking the time to review this manuscript. The authors have considered the comments and tried our best to address them. We hope the manuscript after careful revision meet your standards.

The authors welcome further constructive comments if any.

Below we provide the point-by-point responses. All modifications in the manuscript have been highlighted in red.

 

Comment 1:

The results of the multivariable logistic regression could benefit from a more detailed explanation in the results section. Specifically, further discussion on why hyperactive delirium was associated with better outcomes would strengthen the manuscript.

Response:

We thank the reviewer for this valuable comment and appreciate the opportunity to clarify this issue that caused irritation among many readers of the manuscript.

Unexpectedly, the occurrence of postoperative hyperactive delirium was statistically associated with a good functional outcome. This contradicts the known negative impact of delirium on the prognosis of patients with acute stroke. Previous studies demonstrated that patients with post-stroke delirium have a worse functional outcome and higher mortality. We ascribe the result of our study to the effect of the so-called “survivorship bias”, an error that occurs when researchers focus on groups or cases that have passed a certain selection process. Patients in the “poor functional outcome group” were often in such a bad clinical state, i.e., dead within a few days or unconscious, that they were unable to express symptoms of a delirium. Thus, delirium remained an underrepresented complication in this group. Nevertheless, amid the small sample size of only 8 cases of reported hyperactive delirium, we emphasize that this result must be treated with caution and warrant verification in a larger prospective study.

In order to avoid further confusion or misinterpretation, we added this paragraph to the section Discussion (manuscript page 8, lines 182 - 195).

Comment 2:

The manuscript used Kaplan-Meier analysis for evaluating survival outcomes. Accordingly, Cox proportional hazards models should be more appropriate than logistic regression models for analyzing survival data.

Response:

We thank the reviewer for this hint. However, we solely chose the Kaplan-Meier cumulative survival estimate to visualize the event (death). The main focus of our study is the analysis of the functional outcome measured by the mRS score. Hence, we decided to use logistic regression models.

Comment 3:

The discussion effectively interprets the findings but should provide a more in-depth future plan. For instance, the authors should add the potential to detect penumbral lesions using the advanced MRI technique. The following paper is useful for explaining this technology:

10.3389/fneur.2022.752450

Response:

The mentioned publication is about an imaging study that includes patients with large anterior circulation vessel occlusion and a technology that could help identify the ischemic penumbra in patients with acute ischemic stroke. However, we do not see an appropriate occasion in our manuscript to reference the mentioned publication since we do not consider penumbral lesions of sufficient relevance in the surgical treatment of space-occupying cerebellar infarcts.

Comment 4:

Other MRI techniques can predict the clinical outcome in these patients. Longitudinal changes in iron and myelination within Ischemic lesions detected by QSM were associated with neurological outcomes. Please consider to add it to the Discussion:

10.1161/STROKEAHA.123.044606 

Response:

We thank the reviewer for bringing this interesting publication to our attention. The mentioned study demonstrates the association between longitudinal changes in the magnetic susceptibility values within ischemic lesions and neurological outcomes. The technology used in the study was combined quantitative susceptibility mapping and R2* relaxometry. This might be of interest in future studies about outcome prediction of cerebellar infarction in general. However, in our study, we focussed on the surgical decompressive treatment of large space-occupying cerebellar infarction and its safety and functional outcome. We, therefore, could not find a suitable inducement to link the mentioned work to our study.

 

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

There are no concerns anymore.