Review Reports - Free Flap Reconstruction for Burn Injuries: A Single-Center Retrospective Comparison of Primary and Secondary Cases

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors present a study on free flaps for burn reconstruction, comparing primary and secondary cases. This study is interesting and well-written. However, there are some areas that can be improved:

1) Please indicate the median time between injury and reconstruction. This is interesting and relevant for both groups.

2) How do the groups differ in terms of adjacent procedures (debridement, skin grafting, etc.)?

3) The indications for free flap reconstruction should be explained in greater detail for both groups.

Author Response

Reviewer 1

We thank the reviewer for the positive assessment.

1) Please indicate the median time between injury and reconstruction. This is interesting and relevant for both groups.

We agree, and this information has been added. The date of injury was retrieved for all 65 procedures. In the primary group, free flap coverage was performed a median of 30 days after injury (IQR 22–52; range 9–334), and in the secondary group a median of 534 days (IQR 377–1296), that is, about 1.5 years (range 94 days to 44 years). These figures appear in Table 1 and in Results section 3.2. Because in some secondary cases the injury date was documented only to the month or year, we used mid-point imputation for incomplete dates and report intervals as medians; this is explained in Methods section 2.4 and in the footnote to Table 1. Section 3.2 also notes that the two distributions overlap at their extremes (one primary reconstruction at 334 days for a persistently unhealed wound; one secondary reconstruction at 94 days for an early contracture), reflecting that group assignment was based on the clinical status of the wound rather than a fixed time threshold.

2) How do the groups differ in terms of adjacent procedures (debridement, skin grafting, etc.)?

All operations performed before the free flap were retrieved and are now reported in Table 1 ("Prior surgery before free flap") and in section 3.2. The difference is marked: 33 of 35 primary procedures (94%) had been preceded by debridement only, whereas 26 of 30 secondary procedures (87%) were performed in previously skin-grafted fields and 2 (7%) in fields previously reconstructed with a free flap. Procedures performed after the index free flap (debulking, scar revision, contracture release, additional grafting, flap division) were also retrieved and are reported in the new Table 5 and section 3.6 (please see also our reply to Reviewer 3, comment 4). Debridement, contracture release and skin grafting performed within the same operation as the free flap are included in operative time, which is now defined explicitly in section 2.4.

3) The indications for free flap reconstruction should be explained in greater detail for both groups.

A new subsection (2.2, "Indications for Free Flap Reconstruction") has been added. In brief, in the primary setting a free flap was indicated when, after adequate debridement, a full-thickness burn wound exposed bone, joint, tendon or neurovascular structures that would not reliably support a skin graft, or when a chronic non-healing wound persisted despite conservative treatment; skin grafting alone was the initial treatment for all other acute burn wounds. In the secondary setting, a free flap was indicated for established scar contracture causing functional limitation, unstable or ulcerating scar, or deformity, when the required release created a defect too large for local flaps or when skin grafting was expected to recontract, particularly across joints or in the neck.

Reviewer 2 Report

Comments and Suggestions for Authors

This is a well-written and clinically valuable retrospective study comparing primary and secondary free flap reconstruction in burn patients over a 13-year period. The manuscript effectively highlights the increased operative burden (measured by operative time) of secondary reconstruction in scarred fields despite similar flap complexity. The manuscript is well-structured, clear, and addresses a topic of high relevance in reconstructive microsurgery. However, a few methodological, presentation, and statistical aspects should be addressed. The authors rightly note in the Discussion that the study compares two distinct patient populations (primary vs. secondary flap recipients) rather than comparing early free flap reconstruction to a standard skin-graft-only cohort. Since the vast majority of acute burns are managed successfully with skin grafting alone without developing severe contractures requiring free flaps, the argument for "earlier primary intervention" must be presented strictly as a hypothesis-generating observation. Please ensure this nuance is consistently maintained across the Abstract, Discussion, and Conclusion sections to avoid over-interpreting the findings. The illustrative case figures are helpful, but the watermarks and hospital branding visible in the photographs (e.g., in Fig 1 and Fig 2 panels) should be removed or cropped out to maintain high standards and ensure full anonymization. Please ensure panel labels (A, B, C, D, E, F) in the figure images are uniformly aligned, high-resolution, and clearly referenced within the text description. The extended operative time in secondary cases is attributed to scar excision, contracture release, and recipient vessel isolation in fibrotic fields. As noted in the limitations, step-specific operative times were not recorded. Please briefly elaborate in the Discussion on how future prospective studies could log specific procedural phases (e.g., vessel preparation vs. flap harvesting vs. inset) to confirm this hypothesis empirically.

Author Response

Reviewer 2

This is a well-written and clinically valuable retrospective study comparing primary and secondary free flap reconstruction in burn patients over a 13-year period. … However, a few methodological, presentation, and statistical aspects should be addressed.

We thank the reviewer for the encouraging comments and have addressed each point below.

The authors rightly note in the Discussion that the study compares two distinct patient populations (primary vs. secondary flap recipients) rather than comparing early free flap reconstruction to a standard skin-graft-only cohort. Since the vast majority of acute burns are managed successfully with skin grafting alone without developing severe contractures requiring free flaps, the argument for "earlier primary intervention" must be presented strictly as a hypothesis-generating observation. Please ensure this nuance is consistently maintained across the Abstract, Discussion, and Conclusion sections to avoid over-interpreting the findings.

We fully agree and have revised the manuscript throughout with this in mind. Causal wording such as "attributable to scar-related technical challenges" has been replaced by association language ("associated with a longer operative time … not explained by defect size, burn extent, or flap complexity") in the Abstract, Discussion and Conclusions. The Discussion now opens with a new section 4.1, "Two Distinct Populations", which describes the different clinical pathways of the two groups and states explicitly that the comparison is between two populations and two indications, not between two strategies applied to the same wound. The section on earlier intervention (4.4) has been retitled "The Question of Earlier Intervention" and now states that the study raises that question but does not answer it, and that it is offered strictly as a hypothesis for prospective testing rather than as a recommendation to alter practice; the caveat paragraph on the absence of a skin-graft-only reference group follows immediately. The Abstract conclusion says only that whether earlier coverage could reduce this burden warrants prospective evaluation, and the Conclusions state that the study cannot establish whether earlier free flap coverage would reduce the overall surgical burden. The Limitations repeat that the discussion of expanded indications should be regarded as hypothesis-generating.

The illustrative case figures are helpful, but the watermarks and hospital branding visible in the photographs (e.g., in Fig 1 and Fig 2 panels) should be removed or cropped out to maintain high standards and ensure full anonymization.

Thank you for pointing this out. The marks visible in the original panels were generic text printed on the surgical drapes rather than a hospital name or logo, and did not identify the institution; nevertheless, all panels have been cropped or masked so that no such marks remain, and no clinical content of any image was altered. Revised figures are provided with the resubmission.

Please ensure panel labels (A, B, C, D, E, F) in the figure images are uniformly aligned, high-resolution, and clearly referenced within the text description.

Both figures have been re-assembled with panels of uniform size and labels placed at the same position in the same font. The legends have been rewritten so that each panel is described individually (preoperative, immediate postoperative and final result for each case), and the case descriptions in section 3.7 refer to the panels explicitly.

The extended operative time in secondary cases is attributed to scar excision, contracture release, and recipient vessel isolation in fibrotic fields. As noted in the limitations, step-specific operative times were not recorded. Please briefly elaborate in the Discussion on how future prospective studies could log specific procedural phases (e.g., vessel preparation vs. flap harvesting vs. inset) to confirm this hypothesis empirically.

This has been added to section 4.2, which now closes as follows: "Such studies could resolve the mechanism by logging procedural phases separately — debridement or contracture release, recipient vessel preparation, flap harvest, anastomosis, and inset — and combining phase-specific timing with grading of recipient vessel quality and scar fibrosis." The same section now acknowledges that, because only total operative time was recorded, the relative contribution of these components cannot be separated in the present data, and that contracture release alone, which is not required in primary cases, may account for much of the difference. Phase-specific operative timing is also named as an endpoint for future prospective comparison in the Conclusions.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

Many thanks for submitting your work to JCM!

This is an interesting study with a valuable clinical dataset, but several issues need to be addressed:

-The primary and secondary groups represent very different clinical populations and procedures. The longer operative time cannot be confidently attributed to scar-related dissection based on this comparison alone. This should be presented as an association rather than a causal finding.

-The definition of operative time should be provided. It is unclear whether this included debridement, contracture release, flap harvest, grafting, and donor-site closure.

-Important potential confounders are missing, including burn severity, total body surface area, time from injury to reconstruction, previous operations, recipient-vessel condition, and number of flap components.

-The authors collected data on secondary procedures but did not report them. Total number of operations, revisions, debulking procedures, hospital stay, and long-term functional outcomes would be more meaningful measures of treatment burden than the duration of one operation.

-Please report the exact number of patients overlapping with the previous publication.

-The Data Availability Statement is incomplete, and the duplicated conflict-of-interest statements should be corrected.

Author Response

Reviewer 3

This is an interesting study with a valuable clinical dataset, but several issues need to be addressed.

We thank the reviewer for the thorough review; the comments prompted the additional data collection that has, we believe, considerably strengthened the paper.

The primary and secondary groups represent very different clinical populations and procedures. The longer operative time cannot be confidently attributed to scar-related dissection based on this comparison alone. This should be presented as an association rather than a causal finding.

We agree. All causal wording has been removed (please see also our reply to Reviewer 2, first comment). The finding is now stated in section 4.2 as an association between secondary reconstruction and longer operative time that is not explained by defect size, burn extent or flap complexity, rather than as a demonstration that scar-related dissection is more difficult. The new section 4.1 characterizes the two groups explicitly as distinct populations with different injuries, indications and antecedent treatment, and this is repeated in the Limitations. In addition, the covariates now available (defect size, TBSA, prior operations) were incorporated into the multivariable models so that the alternative explanations could be examined rather than assumed (Table 4, section 3.4).

The definition of operative time should be provided. It is unclear whether this included debridement, contracture release, flap harvest, grafting, and donor-site closure.

A definition has been added to section 2.4: operative time was the total time from the first skin incision to completion of wound closure, and therefore included, as applicable, final debridement or scar excision and contracture release, recipient vessel exposure and preparation, flap harvest, microvascular anastomosis, flap inset, any concurrent skin grafting, and donor site closure; time for anesthetic induction and positioning was not included. Because the durations of individual phases were not recorded, only total operative time was available for analysis.

Important potential confounders are missing, including burn severity, total body surface area, time from injury to reconstruction, previous operations, recipient-vessel condition, and number of flap components.

This comment prompted the record review described at the beginning of this letter. TBSA, burn depth, injury-to-reconstruction interval and all prior operations were retrieved for every case and are reported in Table 1 and section 3.2. TBSA and defect size were added to the multivariable models (Table 4, Models 3–6); the number of flap components is captured by the flap complexity variable (single / flap with graft / chimeric or linked), which was included in Models 2, 5 and 6. In the pre-specified primary model (Model 5: age, sex, defect location, defect size, TBSA and flap complexity), secondary reconstruction remained associated with an adjusted difference of 55.2 minutes (95% CI 20.4–90.1; p = 0.002). Prior operations were not entered as a separate covariate because they are almost perfectly collinear with group membership (94% of primary cases had debridement only, 87% of secondary cases had prior skin grafting) and therefore define, rather than confound, the exposure; they are reported descriptively. Recipient-vessel condition was not systematically documented in the records and could not be retrieved; this is now stated in the Limitations as a source of residual confounding, together with scar quality and the extent of contracture release.

The authors collected data on secondary procedures but did not report them. Total number of operations, revisions, debulking procedures, hospital stay, and long-term functional outcomes would be more meaningful measures of treatment burden than the duration of one operation.

The reviewer is right, and we apologize for the omission. Every procedure performed after the index free flap was retrieved with its date for all patients, and length of the index hospital stay was also extracted. These data are reported in the new Table 5 and the new section 3.6, on a per-flap and a per-patient basis. In brief, at least one secondary procedure followed 40% of primary and 53% of secondary flaps (p = 0.33), with a median of 0 and 1 procedures per flap respectively (p = 0.20); debulking was the most frequent (31% vs 50%; p = 0.14); the number of free flaps per patient (1.2 ± 0.6 in both groups) and the proportion of patients requiring any secondary procedure (47% vs 56%; p = 0.59) were similar. Length of stay was longer in the primary group (median 40 vs 28 days), which reflects the inclusion of acute burn care in the same admission and is annotated as such in Table 1. In section 4.2 we now interpret these findings as indicating that, once free flap reconstruction is undertaken, the downstream refinement burden is comparable regardless of timing, and that the additional burden of secondary reconstruction is concentrated in the index operation. We agree that long-term functional outcomes would be the most meaningful measure; they were not available in this retrospective series, which is stated in the Limitations and identified as a priority for prospective study in the Conclusions.

Please report the exact number of patients overlapping with the previous publication.

The overlap was verified against the patient list of the previous publication (Chang et al., ANZ J Surg 2021). Four patients from the secondary group, treated between 2016 and 2018 for extremity contracture with TDAP flaps, were included in that report; this number is now stated in section 2.1. Most patients of the earlier report were treated before operative time was routinely documented in a retrievable form and were therefore not eligible for the present analysis, which required documented operative time as an inclusion criterion; this is now stated in section 2.1 and acknowledged in the Limitations. The earlier report did not include primary reconstruction cases and did not analyze operative time, and no figure or table is reproduced from it.

The Data Availability Statement is incomplete, and the duplicated conflict-of-interest statements should be corrected.

Thank you for noticing this. The duplicated declarations in the front matter have been removed, and a single set of statements now appears after the Conclusions in the journal's format (Author Contributions, Funding, Institutional Review Board Statement, Informed Consent Statement, Data Availability Statement, Acknowledgments, Conflicts of Interest). The Data Availability Statement now reads: "The data presented in this study are available on request from the corresponding author. The data are not publicly available because they contain information that could compromise the privacy of research participants. The analysis script used to generate the results is likewise available from the corresponding author on request." The IRB statement now includes the protocol code and date of approval.

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors, the manuscript presents an interesting series of free flap reconstructions in burn patients, comparing primary and secondary reconstruction. The series is valuable for this type of surgery and the microsurgical outcomes are excellent.

I have the following comments on the manuscript.

Operative time is the main endpoint and should be better defined. Since the duration of the individual surgical steps is not available, it is difficult to determine how much of the longer operative time is related to dissection through scarred tissue, recipient vessel preparation, or the contracture release itself. I would therefore avoid attributing this difference specifically to the difficulties of dissection in scarred tissue.

The timing of reconstruction is also missing. It would be useful to report the interval between burn injury and free flap reconstruction in the primary group, and between the initial burn and reconstruction in the secondary group. The definition of primary reconstruction as a procedure performed during the same hospitalization is quite broad.

Regarding the multivariable analysis, I would also consider defect size. Defects were larger on average in the secondary group and, given that operative time is the main endpoint, an additional analysis including this variable would be useful.

The methods mention that secondary procedures were collected, but these data are not reported in the results. The number of additional procedures, such as debulking or scar revision, could provide useful information on the overall surgical burden.

I would also suggest updating the discussion with the most recent literature.

the conclusions should more closely reflect what can actually be supported by the study design.

Author Response

Reviewer 4

We thank the reviewer for the kind assessment and the helpful suggestions.

We agree on both points. Operative time is now defined explicitly in section 2.4 (see our reply to Reviewer 3, second comment). The attribution to dissection difficulty has been removed throughout. Section 4.2 now states that the relative contribution of the components of secondary reconstruction cannot be separated in our data, that contracture release alone is a substantive step not required in primary cases and may account for much of the difference, and that the finding is best stated as an association not explained by defect size, burn extent or flap complexity rather than as evidence that scar-related dissection is more difficult. The Abstract and Conclusions were revised in the same spirit.

The interval is now reported for both groups (primary: median 30 days, IQR 22–52, range 9–334; secondary: median 534 days, IQR 377–1296, about 1.5 years, range 94 days to 44 years) in Table 1 and section 3.2. We agree that "same hospitalization" alone is broad. Section 2.1 now specifies that primary reconstruction was performed before wound maturation, for an acute or subacute wound that had not healed with conservative measures or grafting, and that in this series all primary reconstructions were performed within 12 months of injury and 34 of 35 within 100 days; secondary reconstruction was performed after discharge from initial burn care for an established contracture, unstable scar or deformity. Section 3.2 explains that group assignment was based on the clinical status of the wound rather than a fixed time threshold and describes the two cases at the extremes. As noted at the beginning of this letter, this verification also led us to reclassify two cases from primary to secondary.

We are grateful for this suggestion, which turned out to be important: after the reclassification described above, defect size differs significantly between groups (348.5 ± 249.5 vs 203.3 ± 165.4 cm²; p = 0.007). Defect size was therefore added to the models, and a full sequence of models is presented in Table 4 and section 3.4. Adjustment for defect size attenuated the estimate (Model 3: 39.5 minutes; 95% CI 1.1–77.9; p = 0.044), indicating that part of the crude difference reflects the greater extent of resurfacing required after contracture release; with defect size and TBSA together (Model 4) the adjusted difference was 45.7 minutes (p = 0.017), and in the pre-specified primary model that also includes flap complexity (Model 5; R² = 0.59) it was 55.2 minutes (95% CI 20.4–90.1; p = 0.002). Defect size itself was independently associated with operative time (12.0 minutes per 100 cm²; p = 0.006). The estimate was stable with cluster-robust standard errors and in the first-flap-per-patient sensitivity analysis. The partial attenuation by defect size is discussed explicitly in section 4.2.

These data have been retrieved and are now reported in the new Table 5 and section 3.6 and discussed in section 4.2; please see our reply to Reviewer 3, fourth comment, for details.

I would also suggest updating the discussion with the most recent literature.

The Discussion has been updated with two recent systematic reviews: a 2025 review of 396 microsurgical burn reconstructions reporting flap success rates of 92.7% for acute and 95.7% for delayed reconstruction (Fan et al., Microsurgery 2025; reference 20), and a 2024 meta-analysis identifying the timing of reconstruction within the acute period as a determinant of flap loss (Alessandri-Bonetti et al., Plast Reconstr Surg Glob Open 2024; reference 21). Both are cited in section 4.3, and the latter also in section 4.4, where the timing-related risk of flap loss is noted as a further consideration bearing on any move toward earlier intervention.

The conclusions should more closely reflect what can actually be supported by the study design.

The Conclusions have been rewritten. They now state the descriptive findings (distinct populations; an association between secondary reconstruction and roughly 55 minutes longer operative time that persisted after adjustment; similar flap survival and secondary-procedure frequency; reliability of thoracodorsal-based flaps in both settings) and then state explicitly that, because the study compares two distinct patient populations and lacks a skin-graft-only reference group, it cannot establish whether earlier free flap coverage would reduce the overall surgical burden of burn reconstruction. The estimate is positioned as a figure for counseling and planning and as an effect size for a prospective comparison. The Abstract conclusion was shortened correspondingly.

Reviewer 5 Report

Comments and Suggestions for Authors

I was glad to review the manuscript entitled "Free Flap Reconstruction for Burn Injuries: A Comparison of Primary and Secondary Cases". The purpose of this study is to ascertain whether secondary reconstruction carries a higher operative burden and, if so, whether this distinction warrants the expansion of primary free flap covering indications in burn care. The authors concluded that, due to scar-related technical difficulties rather than flap complexity, secondary burn reconstruction required more surgery than primary reconstruction.

1) I would suggest adding the information that this is a single-center retrospective
study in the title of the study

2) Be careful with the correct order of the tables. Table 3 is before Table 2

3) The information given for the anterolateral thigh (ALT) flap is derived from a non-recently published article (reference 5 was published in 2003)

4) I would recommend a quick review of research that assesses the data on contracture recurrence rates and failure in free flap delayed burn repair.

5) Implementation and Validation of Free Flaps in Acute and Reconstructive Burn Care

6) I would recommend a little talk about cases of hydrofluoric acid burn injuries treated with early surgery using free flap repair.

7) According to the literature, does Rosuvastatin accelerate the healing process of partial-thickness burn wounds?

8) Lastly, briefly examine how the timing of microsurgical reconstruction affects the risk of free flap loss in acute burns.

Author Response

Reviewer 5

I was glad to review the manuscript entitled "Free Flap Reconstruction for Burn Injuries: A Comparison of Primary and Secondary Cases". … The authors concluded that, due to scar-related technical difficulties rather than flap complexity, secondary burn reconstruction required more surgery than primary reconstruction.

We thank the reviewer for the careful reading. We would note that, in response to Reviewers 2–4, the conclusion has been reframed: the manuscript no longer attributes the difference to scar-related technical difficulty but reports an association between secondary reconstruction and longer operative time that is not explained by defect size, burn extent or flap complexity.

1) I would suggest adding the information that this is a single-center retrospective study in the title of the study.

Done. The title now reads "Free Flap Reconstruction for Burn Injuries: A Single-Center Retrospective Comparison of Primary and Secondary Cases".

2) Be careful with the correct order of the tables. Table 3 is before Table 2.

Thank you. The Results have been reorganized so that tables are cited in numerical order (Table 1 in sections 3.1–3.2, Tables 2 and 3 in section 3.3, Table 4 in section 3.4, and Table 5 in section 3.6).

3) The information given for the anterolateral thigh (ALT) flap is derived from a non-recently published article (reference 5 was published in 2003).

A recent systematic review of free ALT flaps in lower-limb reconstruction (Reisinger et al., J Clin Med 2026; reference 6) has been added alongside the original citation in the Introduction.

4) I would recommend a quick review of research that assesses the data on contracture recurrence rates and failure in free flap delayed burn repair.

The 2024 systematic review and meta-analysis of free flap failure and contracture recurrence in delayed burn reconstruction (Liu et al.; reference 7) is now cited explicitly in this context in the Outcomes section (4.3), in addition to its citation in the Introduction.

5) Implementation and Validation of Free Flaps in Acute and Reconstructive Burn Care.

We thank the reviewer for drawing our attention to this article (Ziegler et al., Medicina 2021), which is now cited in section 4.3 (reference 21) as a comparator for acute and reconstructive flap outcomes.

6) I would recommend a little talk about cases of hydrofluoric acid burn injuries treated with early surgery using free flap repair.

A sentence has been added to section 4.4 noting that early definitive free flap coverage is already accepted practice in selected chemical injuries such as hydrofluoric acid burns of the hand, where progressive tissue necrosis favors early radical debridement and flap reconstruction (reference 24). This example serves as a precedent within the discussion of earlier intervention, which is raised as a hypothesis for wounds at predictable high risk of contracture.

7) According to the literature, does Rosuvastatin accelerate the healing process of partial-thickness burn wounds?

We are aware of experimental work suggesting that statins, including rosuvastatin, may accelerate the healing of partial-thickness burn wounds in animal models. The present study, however, concerns full-thickness injuries with exposed critical structures and established contractures, in which the question is one of surgical coverage rather than epithelial healing, and we felt that a discussion of pharmacological adjuncts for partial-thickness wounds would fall outside its scope. We hope the reviewer will understand this decision.

8) Lastly, briefly examine how the timing of microsurgical reconstruction affects the risk of free flap loss in acute burns.

This has been added to section 4.3, citing the 2024 meta-analysis of 275 acute burn free flaps (Alessandri-Bonetti et al.; reference 23), which found the highest failure rate for flaps performed 5–21 days after injury (16.6%) compared with 7.3% within 4 days and 6.7% between 22 days and 6 weeks; we note that most primary flaps in the present series were performed beyond this window (median 30 days).

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed all of my comments appropriately.