Effect on timing of free flap breast reconstruction on mastectomy skin necrosis

Authors: Ahmed S, Crabtree J, Fallah KN, Rinne EJ, Hulsman L, Fisher CS, Ludwig KK, Danforth RM, Lester ME, Hassanein AH

Affiliation: Indiana University School of Medicine

Journal: Journal of Reconstructive Microsurgery, 2026

PMID: 40068867

Key takeaways:

• Immediate DIEP had higher mastectomy skin necrosis than delayed-immediate DIEP (11.3% vs 2.2%; p = 0.025).

• Necrosis needing operative debridement was higher with immediate DIEP (7.5% vs 1.1%; p = 0.0499).

• Absolute risk reduction for mastectomy skin necrosis with delayed-immediate approach: 9.1%; NNT ≈ 11 flaps to prevent one necrosis event.

Background: Mastectomy skin necrosis (MSN) remains a meaningful complication after autologous reconstruction, and internal mammary harvest/anastomosis requires prolonged skin-flap retraction that may worsen ischemia in fresh mastectomy flaps.

Objective: Determine whether timing of DIEP reconstruction (immediate vs delayed-immediate) affects mastectomy MSN.

Methods:

• Design: Single-center retrospective comparative study.

• Timeframe: 2013–2016.

• Cohorts:

  • Group I: Immediate DIEP (49 patients, 80 flaps; 31 bilateral).

  • Group II: Delayed-immediate DIEP (57 patients, 93 flaps; 36 bilateral).

• Key exclusions: “Delayed DIEP” with no immediate reconstruction.

• Endpoints:

  • Primary: MSN and management (local care, wound clinic referral, operative debridement).

  • Secondary: DIEP flap skin necrosis, takeback, flap loss, and wound healing complications.

Results:

• Baseline differences:

  • Diabetes mellitus: 13.8% immediate vs 31.2% delayed-immediate (p = 0.001).

  • Adjuvant radiation: 0% immediate vs 64.9% delayed-immediate (p = 0.0001).

  • Periareolar incision more common in delayed-immediate (38.8% vs 61.4%; p = 0.0027).

• Primary outcome (mastectomy skin flap necrosis):

  • 11.3% (9/80) immediate vs 2.2% (2/93) delayed-immediate; p = 0.025.

  • Absolute risk reduction 9.1%; NNT ≈ 11 flaps.

• Management of MSN:

  • Local wound care only: 3.8% (3/80) immediate vs 1.2% (1/93) delayed-immediate; p = 0.3369.

  • Operative debridement/reclosure: 7.5% (6/80) immediate vs 1.1% (1/93) delayed-immediate; p = 0.0499.

• Other flap outcomes (not statistically significant):

  • DIEP flap partial skin necrosis: 5.0% immediate vs 1.1% delayed-immediate; p = 0.183.

  • Takeback: 6.3% immediate vs 4.3% delayed-immediate; p = 0.467.

  • Flap loss: 5.0% immediate vs 1.1% delayed-immediate; p = 0.167.

• Follow-up: Mean 472 days (range 99–1,381).

Conclusion: Immediate DIEP performed on the day of mastectomy had a significantly higher risk of MSN than a delayed-immediate approach, so timing can be used in counseling as a modifiable risk factor.

Strengths:

• Clear, clinically actionable endpoint (MSN and escalation of care).

• Homogenized comparison by excluding “true delayed” DIEP cases with skin deficiency.

Limitations:

• Retrospective, single-center design with modest sample size and few events (risk of residual confounding).

• Cohorts differ meaningfully (e.g., diabetes and adjuvant radiation), and no multivariable adjustment is reported in the methods/statistics section.

• Uses flaps as units for some analyses (possible non-independence in bilateral cases).

Clinical relevance: For patients at higher wound-risk (e.g., diabetes), this paper supports considering a delayed-immediate pathway (TE first, DIEP later) to reduce MSN and the need for operative debridement, at the cost of an additional operation.

Editorial notes:

• The headline finding (MSN reduction) is compelling, but the absence of risk-adjusted modeling is the major weakness. Given major baseline differences (diabetes, incision type, radiation exposure), I’d want:

  • A patient-level multivariable model (or propensity approach) accounting for diabetes, BMI, smoking, mastectomy type/incision, mastectomy weight, prior radiation, and surgeon.

  • Clarification of the definition and adjudication of “mastectomy skin necrosis” (clinical vs imaging, threshold for calling it, standardized follow-up interval).

• The proposed mechanism, retraction stress on acutely ischemic skin in immediate DIEP vs “vascular delay” in delayed-immediate, is plausible and essentially echoes the delay phenomenon, but could’ve directly tested through SPY/ICG perfusion assessment in both cohorts.

• The “NNT ≈ 11 flaps” is clinically useful, but the event count is small. Confidence intervals for ARR/NNT would be helpful.

• Finally, delayed-immediate carries its own hazards (TE infection/removal, expander complications), and those outcomes weren’t quantified here, yet they materially affect counseling.