Optimising Patient Safety During the Use of Intraoperative Imaging for the Surgical Management of Facial Fractures: A Pilot Study

Round 1

Reviewer 1 Report

The study is well structured, and the main finding, significantly lower radiation dose with CBCT compared to conventional CT, is clear and meaningful. The topic is appropriate for publication. However, several methodological issues should be clarified to strengthen the validity of the conclusions.

1. Sample size and missing data

Only 24 out of 48 patients were included in the final analysis

• The reason for excluding half of the cohort is not clearly explained
• This may introduce selection bias

Suggestion:
Please clarify the reasons for missing or excluded data and consider comparing included vs excluded cases if possible.

2. Lack of matched comparison

The CBCT group and the CTFB group are not directly comparable:

• CBCT: surgical patients
• CTFB: 30 “random” patients

This may introduce bias, as imaging indications and protocols may differ.

Suggestion:
This limitation should be more clearly acknowledged in the discussion. A matched or paired comparison would strengthen the study.

3. Variability in CT protocols

The manuscript notes that some CT scans included combined brain and facial imaging

• This likely increases radiation dose in the CT group
• It may exaggerate the difference between CBCT and CT

Suggestion:
Clarify how many cases used each protocol, or discuss this more explicitly as a source of bias.

Author Response

1. Sample size and missing data

Only 24 out of 48 patients were included in the final analysis

• The reason for excluding half of the cohort is not clearly explained
• This may introduce selection bias

Suggestion:
Please clarify the reasons for missing or excluded data and consider comparing included vs excluded cases if possible.

Thank you for this point - the data was not included due to a potential system error, and as such dose and DAP data were not available. Resultantly, these patients were not included in the study. We have also highlighted this in our discussion (please see lines 264-275).

2. Lack of matched comparison

The CBCT group and the CTFB group are not directly comparable:

• CBCT: surgical patients
• CTFB: 30 “random” patients

This may introduce bias, as imaging indications and protocols may differ.

Suggestion:
This limitation should be more clearly acknowledged in the discussion. A matched or paired comparison would strengthen the study.

Thank you for your input. The “random” sampling of these 30 control group patients was convenience sampling to reflect the clinical setting. However, we do acknowledge this as a bias and such have stated more explicitly in the revised manuscript (please see lines 116-117).

3. Variability in CT protocols

The manuscript notes that some CT scans included combined brain and facial imaging

• This likely increases radiation dose in the CT group
• It may exaggerate the difference between CBCT and CT

Suggestion:
Clarify how many cases used each protocol, or discuss this more explicitly as a source of bias.

We acknowledge the scan variability and its implication as a source of bias, and thus have highlighted it more explicitly in the manuscript (please see lines 230-231).

Reviewer 2 Report

1. The Methods don't give a clear description of the control group. There are some important details missing from the comparison, such as the patient selection, the clinical indication, and the scan protocol.
2. Concerns about selection bias have been raised because only 24 of the 48 patients were included (50% were left out because of missing data). What information was missing? Were the patients who were not included different in some way, like the type of fracture or the scan protocol?
3. Important clinical information was not gathered, like the type of fracture, the severity of the injury, and the reason for the imaging. These can change the amount of radiation given (for example, ZMC vs. panfacial fractures, facial CT vs. combined brain + facial CT), so they should be included or made clear as a limitation in discussion.
4. Because of the small sample size and way the study was set up, it might be better to call it an exploratory or pilot study, and the conclusions should be more cautious.

Author Response

1. The Methods don't give a clear description of the control group. There are some important details missing from the comparison, such as the patient selection, the clinical indication, and the scan protocol.

Thank you for your input. The “random” sampling of these 30 control group patients was convenience sampling to reflect the clinical setting. However, we do acknowledge this as a bias and such have stated more explicitly in the revised manuscript (please see lines 116-117).

1. Concerns about selection bias have been raised because only 24 of the 48 patients were included (50% were left out because of missing data). What information was missing? Were the patients who were not included different in some way, like the type of fracture or the scan protocol?

Thank you for this point - the data was not included due to a potential system error, and as such dose and DAP data were not available. Resultantly, these patients were not included in the study. We have also highlighted this in our discussion (please see lines 264-275)

1. Important clinical information was not gathered, like the type of fracture, the severity of the injury, and the reason for the imaging. These can change the amount of radiation given (for example, ZMC vs. panfacial fractures, facial CT vs. combined brain + facial CT), so they should be included or made clear as a limitation in discussion.

Thank you for the interesting point – we acknowledge in the methods that this was not included as this was not a focus of the study; however we also acknowledge that the type of fracture influences the scan and thus the radiation. We have highlighted this in the discussion (please see lines 277-280).

1. Because of the small sample size and way the study was set up, it might be better to call it an exploratory or pilot study, and the conclusions should be more cautious.

We have taken this into account and have thus included this in the title (please see line 3).