Faculty Bibliography

BACKGROUND:Free flap monitoring in autologous reconstruction after nipple-sparing mastectomy (NSM) remains controversial. We therefore examined outcomes in NSM with buried free flap reconstruction versus free flap reconstruction incorporating a monitoring skin paddle. METHODS:Autologous free flap reconstructions with NSM performed from 2006 to 2015 were identified. Demographics and operative results were analyzed and compared between buried flaps and those with a skin paddle for monitoring. RESULTS:221 free flaps for NSM reconstruction were identified: 50 buried flaps and 171 flaps incorporating a skin paddle. Most common flaps used were deep inferior epigastric perforator (DIEP) (64%), profunda artery perforator (PAP) (12.1%), and muscle-sparing transverse rectus abdominis myocutaneous (MS-TRAM) flaps (10.4%). Autologous reconstructions with a skin paddle had significantly greater BMI (p=0.006). Mastectomy weight (p = 0.017) and flap weight (p<0.0001) were significantly greater in flaps incorporating a skin paddle. Comparing outcomes, there were no significant differences in flap failure (2.0% vs. 2.3%, p=1.000) or percentage of flaps requiring return to the operating room (6.0% vs. 4.7%, p=0.715) between groups. Buried flaps had an absolute greater mean number of revisional procedures per NSM (0.82) compared to the skin paddle group (0.44), however rates of revision procedures per NSM were statistically equivalent between the groups (p=0.296). CONCLUSIONS:While buried free flap reconstruction in NSM has been shown to be safe and effective, our technique has evolved to favor incorporating a skin paddle, which allows for clinical monitoring and can be removed at the time of secondary revision.

A retrospective Institutional Review Board-approved review was performed at the Institute of Reconstructive Plastic Surgery, NYU Langone Medical Center, of patients undergoing craniofacial distraction osteogenesis procedures using the rigid external distractor device between 2000 and 2010. Three particularly challenging cases were identified and are presented here.

PURPOSE/OBJECTIVE:Mandibular defects involving the condyle represent a complex reconstructive challenge for restoring proper function of the temporomandibular joint (TMJ) because it requires precise bone graft alignment for full restoration of joint function. The use of computer-aided design and manufacturing (CAD/CAM) technology can aid in accurate reconstruction of mandibular condyle defects with a vascularized free fibula flap without the need for additional adjuncts. The purpose of this study was to analyze clinical and functional outcomes after reconstruction of mandibular condyle defects using only a free fibula graft with the help of virtual surgery techniques. MATERIALS AND METHODS/METHODS:A retrospective review was performed to identify all patients who underwent mandibular reconstruction with only a free fibula flap without any TMJ adjuncts after a total condylectomy. Three-dimensional modeling software was used to plan and execute reconstruction for all patients. RESULTS/RESULTS:From 2009 through 2014, 14 patients underwent reconstruction of mandibular defects involving the condyle with the aid of virtual surgery technology. The average age was 38.7 years (range, 11 to 77 yr). The average follow-up period was 2.6 years (range, 0.8 to 4.2 yr). Flap survival was 100% (N = 14). All patients reported improved facial symmetry, adequate jaw opening, and normal dental occlusion. In addition, they achieved good functional outcomes, including normal intelligible speech and the tolerance of a regular diet with solid foods. Maximal interincisal opening range for all patients was 25 to 38 mm with no lateral deviation or subjective joint pain. No patient had progressive joint hypomobility or condylar migration. One patient had ankylosis, which required release. CONCLUSION/CONCLUSIONS:TMJ reconstruction poses considerable challenges in bone graft alignment for full restoration of joint function. The use of CAD/CAM technology can aid in accurate reconstruction of mandibular condyle defects with a vascularized free fibula flap through precise planning and intraoperative manipulation with optimal functional outcomes.

BACKGROUND: Left ventricular assist devices (LVADs) have become useful adjuncts in the treatment of patients with end-stage heart failure. LVAD implantation is associated with a unique set of problems; one such problem is device infection. We report our experience with flap salvage of infected and/or exposed LVAD hardware. METHODS: Between 2011 and 2016, 49 patients underwent LVAD implantation at our institution. Patients were then categorized by infectious status: systemic infection not directly involving the LVAD device, hardware infection responsive to antibiotics, and exposure of LVAD hardware or device infection refractory to antibiotics requiring debridement and flap coverage. RESULTS: Approximately 50% of device-related infections resolved with either oral or intravenous antibiotics while the other 50% necessitated debridement and coverage with healthy tissue. In total, 12 patients (24%) developed a device-related infection ranging from superficial driveline cellulitis to purulent pocket infections. Seven patients (14%) required extensive debridement and/or flap coverage. CONCLUSION: Early debridement and coverage of exposed hardware are crucial to successfully treating these LVAD infections.

BACKGROUND: Open tibia fractures are commonly stratified by the Gustilo classification, an orthopedic grading system that does not incorporate the presence of arterial injury when limb perfusion is intact. In the authors' experience, however, the presence of arterial injury appears to negatively impact microsurgical outcomes. METHODS: In a retrospective review of 806 lower extremity reconstructions between 1979 and 2016, 361 soft-tissue flaps performed for Gustilo type IIIB/C coverage met inclusion criteria. Patient demographics, flap characteristics, and outcomes were analyzed. RESULTS: Most patients suffered type IIIB [n = 332 (91.9 percent)] injuries; 29 (8.0 percent) had type IIIC injuries. Preoperative angiography [n = 243 (67.3 percent)] demonstrated arterial injury in 126 (51.8 percent); 27 arterial injuries were identified intraoperatively; and the overall incidence was 153 of 361 (42.4 percent). Complications occurred in 143 flaps (39.6 percent) and included 37 partial losses (10.2 percent) and 31 total losses (8.6 percent). Injured recipient arteries [n = 62 (17.2 percent)] had more complications (p = 0.004); specifically, increased take-backs (p = 0.009). Decreasing vessel runoff increased the risk of complications (p = 0.025), take-backs (p = 0.007), and total flap failures (p = 0.024) accordingly. Specifically, among grade IIIB injuries, controlling for age, sex, time since injury, and vein number, single-vessel runoff was associated with higher rates of complications (relative risk, 3.07; p = 0.012), take-backs (relative risk, 3.43; p = 0.013), and total flap failures (relative risk, 4.80; p = 0.010) compared with three-vessel runoff. CONCLUSIONS: Arterial injury was common among Gustilo type IIIB patients and correlated with increased reconstructive complications. Nonischemic arterial injury appears to negatively impact reconstructive outcomes and should be accounted for when considering free tissue transfer for lower extremity salvage. The authors propose a 3-2-1 modification of the Gustilo type IIIB classification to incorporate degree of arterial injury, as it appears to add prognostic value and certainly influences the reconstructive plan. CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, IV.

OBJECTIVE:Driveline infection is a common complication of durable left ventricular assist device support. The majority involve the driveline exit site and can be treated with antibiotics and local wound care. Less frequently, these infections extend into deeper tissues and surgical debridement is necessary. Few studies have described the surgical strategy for treatment of deep driveline infection or have reported long-term outcomes. With a growing population of patients being implanted as destination therapy, there is an obvious need to evaluate and optimize treatment for complex driveline infections. METHODS:Outcomes of patients undergoing durable left ventricular assist device implantation at a single center between 2011 and 2017 were reviewed retrospectively. Data including occurrence of driveline infection, pathogen, time to driveline infection, and treatment strategy were abstracted from the electronic medical record. RESULTS:Driveline infection occurred in 10 (16.4%) of 61 patients at a median of 362 days (Q1 = 99, Q3 = 694) after primary left ventricular assist device implantation. Three (30.0%) of 10 driveline infections were categorized as deep and did not resolve with intravenous antibiotic therapy. In these cases, a multistage approach that included initial debridement and exteriorization of the infected driveline, followed by delayed surgical relocation of the driveline in a clean vascularized soft tissue bed, was used. Long-term device salvage was achieved in all cases. CONCLUSIONS:An aggressive surgical strategy, including debridement and formal relocation of the driveline exit site, can result in long-term device salvage after deep driveline infection. This approach is a less invasive alternative to device exchange for refractory driveline infections.

History of smoking has been implicated as a risk factor for reconstructive complications in nipple-sparing mastectomy (NSM), however there have been no direct analyses of outcomes in smokers and nonsmokers. All patients undergoing NSM at New York University Langone Medical Center from 2006 to 2014 were identified. Outcomes were compared for those with and without a smoking history and stratified by pack-year smoking history and years-to-quitting (YTQ). A total of 543 nipple-sparing mastectomies were performed from 2006 to 2014 with a total of 49 in patients with a history of smoking. Reconstructive outcomes in NSM between those with and without a smoking history were equivalent. Those with a smoking history were not significantly more likely to have mastectomy flap necrosis (p = 0.6251), partial (p = 0.8564), or complete (p = 0.3365) nipple-areola complex (NAC) necrosis. Likewise, active smokers alone did not have a higher risk of complications compared to nonsmokers or those with smoking history. Comparing nonsmokers and those with a less or greater than 10 pack-year smoking history, those with a > 10 pack-year history had significantly more complete NAC necrosis (p = 0.0114, <0.0001). Those with <5 YTQ prior to NSM trended toward an increased rate of complete NAC necrosis (p = 0.0752). Outcomes for those with a < 10 pack-year smoking history or >5 YTQ prior to NSM were equivalent to those without a smoking history. We demonstrate that NSM may be safely offered to those with a smoking history although a > 10 pack-year smoking history or <5 YTQ prior to NSM may impart a higher risk of reconstructive complications, including complete NAC necrosis.

A global outbreak of invasive Mycobacterium chimaera infections after cardiac surgery has recently been linked to bioaerosols from contaminated heater-cooler units. The majority of cases have occurred after valvular surgery or aortic graft surgery and nearly half have resulted in death. To date, infections in patients with left ventricular assist devices (LVADs) have not been characterized in the literature. We report two cases of device-associated M. chimaera infection in patients with continuous-flow LVADs and describe challenges related to diagnosis and management in this population.

BACKGROUND: Patients undergoing implant-based reconstruction in the setting of postmastectomy radiation therapy suffer from increased complications and inferior outcomes compared with those not irradiated, but advances in radiation delivery have allowed for more nuanced therapy. The authors investigated whether these advances impact patient outcomes in implant-based breast reconstruction. METHODS: Retrospective chart review identified all implant-based reconstructions performed at a single institution from November of 2010 to November of 2013. These data were cross-referenced with a registry of patients undergoing breast irradiation. Patient demographics, treatment characteristics, and outcomes were analyzed. RESULTS: Three hundred twenty-six patients (533 reconstructions) were not irradiated, whereas 83 patients (125 reconstructions) received radiation therapy; mean follow-up was 24.7 months versus 26.0 months (p = 0.49). Overall complication rates were higher in the irradiated group (35.2 percent versus 14.4 percent; p < 0.01). Increased maximum radiation doses to the skin were associated with complications (maximum dose to skin, p = 0.05; maximum dose to 1 cc of skin, p = 0.01). Different treatment modalities (e.g., three-dimensional conformal, intensity-modulated, field-in-field, and hybrid techniques) did not impact complication rates. Prone versus supine positioning significantly decreased the maximum skin dose (58.5 Gy versus 61.7 Gy; p = 0.05), although this did not translate to significantly decreased complication rates in analysis of prone versus supine positioning. CONCLUSIONS: As radiation techniques evolve, the maximum dose to skin should be given consideration similar to that for heart and lung dosing, to optimize reconstructive outcomes. Prone positioning significantly decreases the maximum skin dose and trends toward significance in reducing reconstructive complications. With continued study, this may become clinically important. Interdepartmental studies such as this one ensure quality of care. CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.

Background Higher body mass index (BMI) has been shown to increase postoperative complications in autologous breast reconstruction. However, the correlation with flap weight is unknown. Here, we explore the relationship of flap weights and complication rates in patients undergoing microvascular breast reconstruction. Methods Retrospective chart review identified all patients undergoing microvascular breast reconstruction with abdominally based flaps at a single institution between November 2007 and April 2013. Breasts with documented flap weight and 1-year follow-up were included. Patients undergoing stacked deep inferior epigastric perforator flaps were excluded. Breasts were divided into quartiles based on flap weight and examined by demographics, surgical characteristics, complications, and revisions. Results A total of 130 patients undergoing 225 flaps were identified. Patients had a mean age of 50.4 years, mean BMI of 27.1 kg/m2, and mean flap weight of 638.4 g (range: 70-1640 g). Flap weight and BMI were directly correlated. Flaps were divided into weight-based quartiles: first (70-396 g), second (397-615 g), third (616-870 g), and fourth (871-1640 g). There were no associations between flap weight and incidences of venous thrombosis, arterial thrombosis, hematoma, flap loss, fat necrosis, or donor site hernia. However, increased flap weight was associated with increased rate of donor site wound healing problems in both univariate and multivariate analysis. Conclusions Increased flap weight is not associated with added flap complications among patients undergoing microvascular breast reconstruction, however, patients with flaps of 667.5 g or more are more likely to have donor site healing problems. The success and evidence contrary to previous studies may be attributed to surgeon intraoperative flap choice.