Scientific Abstract Presentations: Reconstruction Session 4 Friday, October 16 Fri, Oct 16 5:00-6:00 p.m. CDT

Purpose: With the global rise of diabetes, diabetic foot ulcers(DFUs) represent a leading cause of chronic, refractory wounds. While randomized controlled trials(RCTs) have evaluated biological substitutes for DFUs, reported clinical benefits remain inconsistent. Compared with the standard of care(SOC) – debridement, moist dressings, and pressure offloading – biologics introduce substantial costs. Importantly, statistical significance alone does not reflect the robustness of treatment effects. Thus, statistical fragility analysis can assess the strength of outcomes by quantifying how easily significance is overturned. This study aimed to evaluate the statistical fragility of RCT outcomes to determine whether biologics offer benefit over SOC in DFU management.

Methods: Following PRISMA guidelines, PubMed, Scopus, CENTRAL, and Web of Science were searched for RCTs evaluating biological substitutes for DFUs. Extracted outcomes included complete wound closure, adverse events, and recurrence. Fragility was assessed using Fisher's exact test with iterative changes to the proportion of outcome events. The fragility index(FI) represented the number of outcome changes needed to reverse statistical significance, while the reverse fragility index(rFI) measured the number needed to overturn nonsignificance. Fragility and reverse fragility quotients (FQ, rFQ) divide FI/rFI by the sample size to standardize outcomes, where an FQ of 0.05 indicates that five outcome changes per 100 patients would alter significance. Data was reported as Median(Q1-Q3).

Results: Thirty-three articles(N=3,459) were included, with 1,302 patients treated with dermal-matrices, 606 with placental membranes, 163 with epidermal-matrices, and 545 controls receiving SOC. Only 16(48.48%) trials performed power analyses, and a mean of 2.5(range: 0-24) patients per study were lost to follow-up. Biologics demonstrated superiority for wound closure in 13 of 23 trials, with a median FI of 6(3-8) and FQ of 0.082(0.024-0.130). Nonsignificant studies had median rFI of 4(3-4) and rFQ of 0.105(0.103-0.139). Meta-analysis favored dermal-matrices (OR[95% CI], 1.61[1.26-2.05], p<0.001; FI=56; FQ=0.047) and amniotic-membranes (1.95[1.53-2.49], p<0.001; FI=88; FQ=0.105) over SOC. Fragility quotients of amniotic membrane trials demonstrate comparatively greater robustness than dermal matrices. Studies directly comparing amniotic membranes and dermal matrices reported mixed, highly fragile wound-closure outcomes, with a mean FI of 3(2-4). Similarly, meta-analysis yielded comparable results(p=0.09; rFI=1; rFQ=0.005) with very high fragility. Adverse events were lower with biologics(26.4%) than SOC(35.5%, p<0.001), but results were not robust(FQ=0.024). No significant difference in disease recurrence was found between biologics and SOC(p=0.192, rFQ=0.036).

Conclusion: RCTs evaluating biologics for DFU treatment demonstrate considerable fragility of outcomes. Across all trials, reversal of statistical significance would occur if only 8.2% of patients treated with biologics failed to achieve wound closure. In multiple studies, FI values were lower than the number of patients lost to follow-up, further highlighting the potential vulnerability of trial findings. Although meta-analyses show improved wound healing with biologics, fragility analysis revealed minimal robustness, with ten or fewer outcome changes per 100 patients overturning significance. Even when comparing amniotic membranes and dermal matrices, data remains fragile, highlighting the need for stronger evidence comparing the cohorts. Given the global burden of diabetic foot ulcers and the considerable cost of biologics, adequately powered trials with robust statistical findings are essential for clinical adoption.

Background: Surgical management of lower extremity tumors has shifted from amputation to limb salvage. Plastic surgery involvement in these procedures may be associated with improved postoperative outcomes. This study further explores how plastic surgery involvement in lower extremity tumor resection correlates with clinical features and postoperative outcomes.

Methods: We conducted a retrospective cohort study of patients who underwent lower extremity tumor resection at a single institution between 2016 and 2024. Demographic, clinical, oncologic, and postoperative characteristics were compared between cases with and without plastic surgery involvement. Predictors of postoperative complications were identified using a backwards stepwise multivariable logistic regression model, with statistical significance defined as p<0.05.

Results: The final cohort included 118 resection cases with median follow up of 8.0 months, 52 (44.1%) of which involved the plastic surgery service. Pedicled or free flap reconstruction was performed in 29 (24.6%) cases, most commonly using gastrocnemius (13, 11.0%) and rectus femoris (4, 3.4%) muscle flaps. Cases with plastic surgery involvement were associated with significantly higher proportions of sarcomas (80.8% vs. 51.5%, p=0.0019), preoperative radiation (28.8% vs. 12.1%, p=0.041), and resection of recurrent tumors (26.9% vs. 10.6%, p=0.040) when compared to cases without plastic surgery involvement. Rates of wound-related complications (25.0% vs. 13.6%, p=0.18), delayed amputation (1.9% vs. 4.5%, p=0.79), local recurrence (9.6% vs. 6.1%, p=0.71), and death (17.3% vs. 19.7%, p=0.93) were additionally not significantly different in the two groups. Preoperative radiation was independently associated with higher odds of wound-related complications (OR: 3.00, 95% CI: 1.01-8.96, p=0.049), and plastic surgery involvement was not a significant predictor (OR: 1.62, 95% CI: 0.58-4.50, p=0.36).

Conclusion: Plastic surgery services were more likely to be involved in the resection of recurrent, irradiated, and/or malignant lower extremity tumors. Despite higher rates of oncologic severity and comorbidities, cases with plastic surgery involvement had no significant differences in postoperative complications compared to cases managed without plastic surgery. These findings support a multidisciplinary approach in the management of complex lower extremity tumors.

Introduction: Motorcycle-related "road rash" often creates broad superficial partial-thickness friction burns. While split-thickness skin grafting (STSG) is the current gold standard of management, negative outcomes are present, including graft rejection, infection, donor-site morbidity, amongst others. One alternative to STSG is autologous skin cell suspensions (ASCS), which has been recognized for their ability to decrease postoperative pain, have comparable cosmetic results, while requiring less donor-site morbidity.

Methods: Three motorcycle crash patients underwent operative debridement followed by ASCS application on post-trauma days 2-5 for partial-thickness wounds on their extremities. STSG was reserved for focal deeper injuries when indicated and each ASCS application used a 24 cm2 donor harvest. One patient (12% TBSA) received ASCS to ~390 cm2 across bilateral forearms, hands, and knees. A second patient (23% TBSA) received ASCS to bilateral lower extremities with adjunct STSG to select sites totaling 60 cm2. A third patient (16% TBSA, mixed depth) underwent excisional preparation (380 cm2) with STSG to the left elbow and ASCS to the remaining partial-thickness extremity wounds.

Results: Postoperative complications and early outpatient healing were assessed clinically and photographically. Preliminary results show no postoperative complications with ASCS-treated wounds between one and two-month follow-ups, with minimal residual scabbing in one case. Planned re-evaluation at 12-month follow-up.

Conclusion: ASCS was a viable option for extensive traumatic road rash, achieving uncomplicated early healing with small donor harvests, and limiting the amount of STSG required. When approaching patients who have suffered partial-thickness burns on their extremities, ASCS can be considered as a potential first line therapeutic option.

PURPOSE: Frailty represents a multidimensional state of physiologic vulnerability arising from cumulative declines in homeostatic, metabolic, and immune function. Unlike chronological age, frailty better captures the biological aging process that predisposes older adults to impaired wound repair and surgical complications. Full-thickness skin grafts (FTSGs) are frequently performed in this population for reconstruction after oncologic excision, trauma, or chronic wounds, yet the impact of frailty on graft success remains undefined. This study leverages a national database to quantify how graded frailty, measured by the modified frailty index (mFI-5), predicts postoperative morbidity following FTSG, establishing a framework for integrating geriatric vulnerability into reconstructive decision-making.

METHODS: A retrospective cohort analysis was performed using the TriNetX US Collaborative Network, encompassing de-identified electronic health records from 72 healthcare organizations (131 million patients). Patients ≥65 years who underwent FTSG (2005–2025) were identified by CPT codes. Frailty was calculated using the validated mFI-5, which includes hypertension, diabetes mellitus, chronic obstructive pulmonary disease or pneumonia, congestive heart failure, and dependence on care. Patients were categorized as nonfrail (0), prefrail (1), frail (2), or severely frail (≥3). Propensity score matching adjusted for demographic and clinical covariates. Logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CIs) and a statistical significance of p<0.05. Postoperative complications within 60 days, including infection, wound healing problems, hematoma or seroma, scarring, debridement, and graft failure, were identified via ICD-10 and CPT codes.

RESULTS: After matching, 7,810 prefrail, 5,445 frail, and 2,008 severely frail patients were analyzed. Increasing frailty scores correlated with progressively higher overall complication risk: prefrail (OR = 1.12; 95%CI 1.01–1.25; p<0.05), frail (OR = 1.37; 95%CI 1.20–1.56; p<0.0001), and severely frail (OR = 1.51; 95%CI 1.25–1.84; p<0.0001). Infection (OR = 1.98; 95%CI 1.32–2.95; p<0.001) and need for debridement (OR = 2.36; 95%CI 1.28–4.32; p<0.01) were nearly doubled in severely frail patients. Patients classified as frail or severely frail (mFI≥2) had a markedly greater risk for skin graft failure (OR = 2.91; 95%CI 1.47–5.79; p<0.01). Similarly, prefrail patients were also at an increased risk for failure (OR = 2.60; 95%CI 1.48–4.55; p<0.001). Pathologic scarring or hematoma were not consistently associated with frailty.

CONCLUSIONS: Frailty is a biologically meaningful predictor of reconstructive outcomes after FTSG in older adults, exerting a graded effect on infection, graft failure, and reoperation risk. These findings demonstrate that physiologic vulnerability, rather than chronological age alone, governs tissue repair and graft integration. Importantly, frailty represents a potentially modifiable state reflecting impaired immune function, vascular reserve, and regenerative capacity. Even modest frailty confers vulnerability equivalent to or exceeding traditional comorbidity burden. Integrating frailty screening (mFI-5) into preoperative workflows provides an opportunity not only for risk stratification but also for targeted intervention through prehabilitation, medical optimization, and individualized reconstructive planning. Reframing frailty as a modifiable determinant of surgical resilience shifts the paradigm from passive risk recognition to proactive optimization, with potential to improve graft survival and postoperative outcomes in the growing population of older surgical patients.

Background The ideal donor tissue for pharyngoesophageal reconstruction remains a subject of debate, primarily centered on the choice between enteral (free jejunum) and fasciocutaneous (anterolateral thigh or radial forearm) free flaps. In this study we compare surgical and functional outcomes between these two choices for laryngopharyngectomy defect reconstruction.

Methods A retrospective cohort study of 28 patients undergoing free flap pharyngeal reconstruction (July 2013 – April 2021) was performed. Patients were stratified into enteral (n=11) and fasciocutaneous (n=17) cohorts. Primary endpoints included flap survival, early and late complications, hospital length of stay, and nutritional autonomy (final diet and tube-feed dependence).

Results Baseline demographics and radiation prevalence (89.3%) were comparable (p>0.05). The jejunal cohort consisted entirely of total circumferential defects (100%), whereas the fasciocutaneous group comprised 58.8% partial defects (p<0.001). Comparing the two cohorts, flap survival, mean length of stay, complication rates were statistically equivalent (p>0.05). Patients in the enteral cohort were significantly more likely to achieve a regular diet (82% vs 18%, p=0.0066). On multivariable analysis adjusting for radiation, jejunal reconstruction was associated with significantly lower odds of long-term tube-feed dependence (OR 0.15; 95% CI, 0.02–0.84; p=0.0454) as well as higher odds of achieving a regular diet (OR 21.00; 95% CI, 3.44–202.00; p=0.0026).

Conclusion Enteral reconstruction is associated with favorable restoration of nutritional autonomy, even in more extensive pharyngeal defects. These findings suggest that intrinsic properties of the jejunum may confer advantages in swallowing recovery and may be considered particularly in a population where these functional outcomes significantly impact quality of life.

Background: Per National Comprehensive Cancer Network (NCCN) guidelines, sentinel lymph node biopsy (SLNB) is recommended in the treatment of malignant melanoma with Breslow thickness ≥ 0.8 mm or <0.8 mm with ulceration or other high-risk features, to provide accurate staging, guide prognosis, and guide adjuvant therapy. Technetium-99m (Tc-99m) radiocolloid mapping remains the gold standard localization modality. Indocyanine green (ICG) fluorescence enables dynamic intra-operative lymphatic mapping and may improve detection in anatomically complex basins, particularly in the head and neck. We evaluated basin-level concordance and incremental detection using a dual-tracer SLNB protocol.

Methods: A retrospective case series of 25 consecutive patients undergoing SLNB for primary melanoma at a single institution was performed. Standard dual-tracer protocol included preoperative Tc-99m lymphoscintigraphy followed by intraoperative gamma probe localization and intradermal ICG injection with near-infrared fluorescence imaging. Outcomes were analyzed at the nodal basin level and included detection rates, inter-modality concordance, discordant node identification, total sentinel nodes retrieved, operative time, nodal positivity, and complications.

Results: Twenty-five patients comprising 29 nodal basins underwent SLNB, including axillary (n = 15), cervical (n = 6), inguinal (n = 4), parotid (n = 3), and popliteal (n = 1) regions. Median Breslow thickness was 1.0 mm (IQR 0.84-1.71). Twenty-eight basins underwent dual-tracer mapping, and one cervical basin underwent fluorescence-only mapping. Dual-tracer mapping achieved 96.4% basin-level concordance (27/28 basins). In a patient with a primary melanoma of the superior scalp, fluorescence identified two sentinel lymph nodes within a cervical basin that were not detected on Tc-99m lymphoscintigraphy or gamma probe localization. Both nodes were pathologically negative. Several dual-detected nodes demonstrated low gamma counts (<200), in which fluorescence provided real-time visual confirmation. Dual-tracer detection was 100% across all other basins and consistent across BMI subgroups (≤ 30 and ≥ 30). In a separate fluorescence-only cervical case involving a central malar melanoma, ICG localized two sentinel lymph nodes, including one node positive for metastasis. Mean operative time was 90.8 ± 39.7 minutes, with longer duration observed in head and neck basins compared to axillary basins. Overall sentinel node metastasis was identified in 24% of patients. No ICG-related complications occurred.

Conclusions: Dual-tracer SLNB demonstrated high basin-level concordance across anatomic regions and BMI subgroups. In a complex cervical basin, fluorescence identified additional sentinel nodes not detected by Tc-99m, and fluorescence-only mapping successfully localized a metastatic sentinel node in a separate case. These findings suggest fluorescence imaging provides meaningful anatomic guidance in technically challenging head and neck basins while maintaining established oncologic staging standards.

Introduction: While studies demonstrate higher risk of certain post-surgical complications, [1] others suggest that GLP-1-associated weight loss may confer beneficial effects on wound complications following free flap-based complex reconstruction [2]. This large database study sought to 1) determine whether GLP-1s confer higher risk of acute, anesthesia-related, and long-term complications following DIEP flap procedures and 2) understand the potential role that diagnosis of diabetes at surgery may play in these outcome differences.

Methods: A retrospective cohort study utilizing the TriNetX Research Network identified adults who underwent free flap-based complex reconstructive procedures with or without preoperative GLP-1 use. Each cohort was stratified into whether a user or non-user carried a diagnosis of diabetes at index surgery. Propensity-score matching controlled for demographics and co-morbidities, including malignant neoplasms, irradiation, and BMI. Acute outcomes (dehiscence, infection, etc.) were compared if they occurred within 90 days, anesthesia-related outcomes (respiratory insufficiency, hemodynamic instability, etc.) if they occurred within 1 week, and long-term outcomes (revision surgery, abdominal wall hernia, etc.) if they occurred within 5 years following surgery.

Results: 32,467 patients without GLP-1 usage and 704 patients with GLP-1 usage were identified. After matching, there were 697 in each cohort for acute outcomes, 676 for anesthesia-related outcomes, and 676 for long-term outcomes. Compared to non-users, GLP-1 users demonstrated significantly higher rates of flap failure/loss within 90 days (OR=1.472; 95% CI, 1.088-1.992; p=0.0119), but lower rates of respiratory insufficiency (OR=0.624; 95% CI, 0.453-0.86; p=0.0038) and cardiac complications (OR=0.557; 95% CI, 0.364-0.852; p=0.0064) within 1 week, as well as abdominal wall hernia occurrence within 5 years (OR=0.459; 95% CI, 0.259-0.813; p=0.0063). When comparing only diabetic free flap patients with or without GLP-1s, GLP-1 users demonstrated significantly higher rates of flap failure/loss (OR=1.631; 95% CI, 1.146-2.319; p=0.0062) and lower rates of cardiac complications (OR=0.513; 95% CI, 0.312-0.843; p=0.0075), but no other significant differences in any other measured outcomes across various timeframes.

Conclusions: Although preoperative GLP-1 usage may result in higher risk of 90-day flap failure or loss rates, they may confer advantages in terms of certain anesthesia related adverse outcomes or 5-year abdominal wall hernia occurrence. However, presence or absence of perioperative diabetes may impact risk differences, particularly in long-term outcomes following complex reconstructive surgery. Further research on GLP-1s in reconstructive surgery is warranted to investigate the role of other relevant factors, such as malnutrition, elevated BMI, and timing of GLP-1 usage relative to index surgery.

References: [1] Lee, et al. Impact of GLP-1 agonist on surgical wound complications following plastic and reconstructive surgery: A propensity matched cohort large database analysis. Plast Reconstr Surg 2025. [2] Sidhu, et al. Effect of glucagon-like peptide-1 agonists on deep inferior epigastric perforator flap breast reconstruction outcomes in obese patients: A study of 5618 patients. J Plast Reconstr Aesthet Surg 2025.

Background: Plastic surgery involvement for soft-tissue reconstruction in spinal fusion is traditionally reserved for high-risk patients, yet objective criteria guiding this surgical decision-making remains poorly defined.1 Our group previously developed an NSQIP-derived risk scoring system to identify spinal fusion patients at elevated risk for postoperative complications. This study aims to validate this risk calculator and identify the strongest predictors of complications following spinal fusion.

Methods: We performed a retrospective review of patients undergoing spinal fusion (2015–2024) at our institution. Demographics, comorbidities, operative details, and ≥1-year outcomes were extracted from the electronic medical record. Risk scores were calculated using our previously designed calculator: (1 point: steroid use, bleeding disorder, thoracic/lumbar fusion; 2 points: ASA III, BMI >33, operative time >200 min, cervical fusion; 3 points: emergency surgery, ASA IV, dialysis, occipital fusion; 4 points: contaminated wound, ≥13 levels fused). Model performance was assessed via ROC/AUC with Youden index (YI), logistic regression, and negative binomial regression.

Results: 410 patients were included (40% male; mean age 58.8±11.8 years; mean BMI 32.5±5.8 kg/m²; mean risk score 5.1±2.4). Medical complications included pulmonary events (13.4%), renal failure (11.2%), UTI (8.8%), VTE (7.6%), and septic shock (7.6%). Surgical complications included wound dehiscence (12.7%), seroma (11.2%), deep SSI (8.8%), and hardware failure (7.8%). Reoperation and readmission occurred in 18.7% and 28.0% of patients, respectively. Thirty day and one-year mortality were 2.4% and 5.1%, respectively. The model performed best for predicting one-year mortality (AUC=0.70); performed moderately for predicting 30-day mortality (AUC=0.66), medical complications (AUC=0.60), and reoperation (AUC=0.62); and performed poorly for predicting surgical complications (AUC=0.47) and readmission (AUC=0.50). Higher risk score was significantly associated with medical complications (OR 1.16, p<0.001), one-year mortality (OR 1.32, p<0.001), VTE (OR 1.18, p=0.017), pulmonary (OR 1.18, p=0.003), cardiovascular (OR 1.25, p=0.005), and renal complications (OR 1.23, p<0.001), but not surgical or wound-related complications.

Conclusion: Our risk scoring system effectively predicts medical morbidity and mortality following spinal fusion but does not independently predict surgical or wound-related complications. Future work will examine whether risk-stratified plastic surgery involvement improves outcomes in medically high-risk patients.

1. Epstein NE. When does a spinal surgeon need a plastic surgeon? Surg Neurol Int. 2013 May 6;4(Suppl 5):S299-300. doi: 10.4103/2152-7806.111431. PMID: 23878764; PMCID: PMC3716009.

Background:

Chronic pain following mastectomy remains a significant source of long-term morbidity for breast cancer survivors. Although breast reconstruction provides important restorative benefits, its impact on persistent pain remains uncertain, particularly across different reconstructive pathways. We therefore evaluated mastodynia and neuropathic medication utilization across reconstructive modalities.

Methods:

Women undergoing mastectomy for breast cancer were identified (n=69,326) in the TriNetX multi-institutional database. Cohorts included mastectomy alone, immediate implant-based reconstruction, immediate autologous reconstruction, direct-to-implant (DTI), and staged tissue expander–to–implant (TE) reconstruction. Outcomes were indexed at the time of implant placement or reconstruction. Propensity score matching was performed for demographic and oncologic covariates. Outcomes assessed ≥90 days after index included mastodynia and neuropathic medication utilization (gabapentin and pregabalin).

Results:

After propensity score matching, both implant-based and autologous reconstruction were associated with significantly increased mastodynia and neuropathic pain medication use. Implant-based reconstruction demonstrated higher rates of mastodynia compared with mastectomy alone (12.1% vs. 5.4%; RR 2.24, 95% CI 2.07–2.42; p<0.001), as well as increased neuropathic medication utilization (23.1% vs. 16.4%; RR 1.41, 95% CI 1.34–1.48; p<0.001). Similarly, autologous reconstruction was associated with higher rates of mastodynia (11.4% vs. 5.0%; RR 2.28, 95% CI 1.88–2.77; p<0.001) and neuropathic medication use (23.9% vs. 17.7%; RR 1.35, 95% CI 1.21–1.50; p<0.001) relative to mastectomy alone.

Within implant-based reconstruction, direct-to-implant (DTI) and staged tissue expander (TE) approaches demonstrated comparable rates of mastodynia (9.9% vs. 10.8%; RR 1.10, 95% CI 0.87–1.37; p=0.433) and neuropathic medication utilization (18.4% vs. 18.9%; RR 1.03, 95% CI 0.87–1.22; p=0.745), with no statistically significant differences observed between reconstruction strategies.

Conclusion:

Both implant-based and autologous breast reconstruction are associated with significantly increased mastodynia and neuropathic medication utilization compared with mastectomy alone. However, within implant-based reconstruction, implant strategy (direct-to-implant versus staged tissue expander–to–implant) does not significantly influence chronic mastodynia or neuropathic medication use. These findings suggest that reconstruction status itself, rather than implant staging technique, is associated with chronic sensory morbidity following mastectomy. Efforts to mitigate long-term pain should therefore focus on perioperative and patient risk factors rather than reconstruction strategy alone.

INTRODUCTION: Infection often accompanies pelvic pressure injuries (PIs) among those with spinal cord injuries (SCIs). In addition to the location, these PIs tend to be challenging to treat considering their polymicrobial nature from both endogenous skin and fecal flora. A two-stage flap reconstruction technique (surgical debridement with 1-week interim, followed by flap reconstruction) aims to address this challenge, but evidence is inconclusive regarding improved outcomes compared with a single-stage approach due to selection bias (1). This study assessed the outcomes of two-stage flap reconstruction in patients with SCI, correlating them with varying degrees of semi-quantitative microbiological burden and reconstructive complexity.

METHODS: We conducted a retrospective chart review over a 7.5-year period. Individuals with SCI who underwent two-stage flap reconstruction for stage III/IV PIs were analyzed (N=146). The primary outcome was complications in the 90-day postoperative period. Microbiological burden was determined based on semi-quantitative culture analyses ("rare"/"few"= low burden;"moderate"/"many" = high burden) at the time of debridement. Descriptive statistics were performed with stratified analyses when appropriate. Univariable logistic regressions were conducted followed by multivariable logistic regressions. Age at the time of surgery, number of flaps, previous flap reconstruction at the same site, microbiology burden, and malnutrition were selected a priori as covariates to assess predictors of minor (Clavien-Dindo Class I and II), major (Clavien-Dindo Class III, IV, V), and overall complications.

RESULTS: Among 146 patients (mean age: 48.1±14.9 years; 81.5% male), 34.9% had at least one major (n=26, 17.8%) or minor (n=33, 22.6%) complication. Descriptive stratified analyses revealed a step-wise increase in complication rates in parallel with a greater number of flaps across all microbiological burden categories, with rates reaching 100% in patients with high microbiological burden reconstructed with 3 flaps. On univariate analysis, previous flap reconstruction (OR: 3.26-5.61), malnutrition (OR: 2.27-2.79), number of flaps (OR: 2.04-2.38), and reconstruction with 3 flaps (OR: 4.88-6.06) were significant. High microbiology burden was significantly associated with minor complications (OR=2.37). After adjustment, previous flap reconstruction was an independent predictor of overall (aOR=7.06, 95%CI:2.32-21.45), major (aOR=3.66, 95% CI:1.44-9.28) and minor (aOR=3.39, 95%CI:1.37-8.40) complications. Malnutrition was an independent predictor of overall (aOR=2.60, 95% CI:1.13-5.98) and major (aOR=3.09, 95%CI:1.22-7.86) complications.

CONCLUSION: Despite the initial association between complication rates and PI wound bed complexity, outcomes were comparable to previously published studies utilizing single-stage flap reconstruction. In addition, aligned with prior literature, previous flap reconstruction and malnutrition resulted as the strongest predictors, underscoring the efficacy of a two-stage approach to manage complex PIs complicated by various levels of polymicrobial colonization or infection.

Mishra A, Eldolify M, Shirley R, Chan JKK. Flap reconstruction of pressure ulcers in patients with spinal cord injury: a retrospective cohort study. Spinal Cord. 2025;63(6):292-297. doi:10.1038/s41393-025-01080-2