Microvascular Reconstruction: Osteocutaneous Flaps and Complications

Hi Team,

Second week of microvascular reconstruction, and we're moving from soft tissue into bone. Wednesday mornings at 7 AM for our discussion - I'm gonna bring in a Challah and some toppings for anyone who can join in person. Grab any of us in the OR if you can't make the 7 AM session.

This week's topic is osteocutaneous flap selection and the complications that keep us up at night (although sleep deprivation tends to help). We'll discuss plate exposure, malunion, and the perioperative decisions that either prevent those problems or invite them. If you haven't looked through the "Flaps and Reconstructive Surgery" textbook in the shared folder yet by Fu Chan Wei, the bone flap chapters are worthwhile although there could be some more scapula related content in future editions as it's one of the few things that is not well covered in this textbook.

Choosing a Bone Flap: Fibula, Scapula Tip, and Smaller Options

The donor site decision comes down to the same handful of questions every time: how much bone do you need, how many osteotomies will the defect require, does the soft tissue paddle need to move independently of the bone (a chimeric flap), and what can the patient's vasculature and comorbidities actually tolerate. A segmental mandibulectomy for osteoradionecrosis is a different problem than a composite resection for a T4 SCCa with a large skin defect and/or large intraoral defect.

The fibula remains the workhorse for good reason. A systematic review spanning 25 years and nearly 9,500 mandibular defects found fibula, iliac crest, osteocutaneous radial forearm, and scapula were the flaps actually used at scale, and the iliac crest had a meaningfully higher failure rate (6.2%) than fibula, radial, and scapular flaps combined (3.4%) (Brown Br J Oral Maxillofac Surg 2017). Iliac bone crest can result in a hernia and is also immensely painful to the patients which is less well documented in the literature (in case you were wondering why we don't do more of these) though it does have excellent bone stock and a shape that more appropriately matches certain regions of the mandible.

The major alternative to the fibula is the scapula. I tend to favor the lateral border scapula over the scapula tip as it has a more reliable anatomy and blood supply and is less bulky then a combined lat muscle/skin paddle with its fasciocutaneous parascapular skin flap. The main negative for a lateral border is the short pedicle (~4ish cm). The scapula tip flap is a strong choice for short mandible defects or with a large soft tissue requirement or for midface reconstruction since the bone and the latissimus or serratus paddle can move independently off the same pedicle; a series of 120 patients reported 95% met criteria for dental implants afterward, with low donor site morbidity although I personally find the scap tip to have weaker bone than the lateral border and the fibula (Blumberg Head Neck 2019). The osteocutaneous radial forearm free flap (OCRFFF) is worth keeping in your back pocket too, particularly for shorter, anterior arch defects where the fibula's segmental length is overkill and the thinner, pliable soft tissue paddle is actually an asset. It doesn't come free of trade-offs though: a direct comparison to fibula for oncologic mandible reconstruction found significantly more late complications with OCRFFF, including nonunion in 10.7% of cases versus 0% for fibula, and partial or complete flap failure in 7.1% versus 0% (Le Microsurgery 2021). I reserve it for shorter defects where I'm not asking the bone to do much heavy lifting.

Whatever donor site you land on, the preoperative vascular workup is not optional. For fibula specifically, that means lower extremity MRA, CTA, or Doppler evaluation confirming three-vessel runoff (anterior tibial, posterior tibial, peroneal), plus an assessment of foot neurovascular status before you commit to sacrificing the peroneal artery. Skipping this step because a patient "looks fine" is how you end up with an ischemic foot in the recovery room. I prefer CTA as it allows for VSP but if there's any question of vascular compromise / prior injury / edema then don't hesitate to move to a scapula.

Virtual Surgical Planning: How Accurate Is It, Really?

Virtual surgical planning and CAD/CAM cutting guides have become close to standard for mandibular reconstruction, and it's tempting to treat the plan as gospel once it's printed. The honest answer to "how accurate is it" depends on which outcome you're actually measuring. On pure geometric accuracy, a study comparing actual postoperative position against the virtual plan found a median deviation of just over 1 mm overall, but the split by case type matters: about 0.8 mm for alloplastic-only reconstructions versus 2.5 mm for microvascular osseous reconstructions (Mascha J Craniomaxillofac Surg 2017). On the outcomes that actually change how a case goes, a retrospective review of 126 patients comparing customized, VSP-designed plates to traditional intraoperative plate bending found the customized plates cut total OR time (643.0 vs. 741.7 minutes) and trended toward fewer hardware complications (5.5% vs. 15.5%), and independently predicted shorter OR time, fewer overall complications, and shorter length of stay on multivariate analysis (Lee Plast Reconstr Surg 2020). That's the number I'd lean on if a resident asks me to justify the extra planning time and cost. And the cost question itself has its own literature now: a systematic review of 18 studies found that 78% showed VSP either generated cost savings or was cost-comparable to freehand surgery, with the biggest savings tied to reduced OR time and length of stay, and better economics when planning is kept in-house rather than outsourced to a vendor (Xiao Head Neck 2024). I find that a lot of these economics papers don't take the full picture into account though (the cost of hiring an engineer, time required to make the plans yourself instead of with a technician who helps you make the plans in ~20-30 minutes, composite versus nylon versus titanium guides, etc). The literature on VSP is still heavier on OR-time and complication data than on long-term functional outcomes, speech, swallowing, and dental rehabilitation, so don't let a beautifully accurate plan substitute for an honest conversation with the patient about what "success" looks like for them (Baecher Front Med (Lausanne) 2023). My noncontroversial comment would be that VSP is immensely valuable for complex defects where it can be near impossible to get the patient into occlusion (large destructive defects or patients with chronic fracture/malocclusion) and for Jaw-in-a-Day type procedures but less useful for a single segment defect that can be fixated/plated prior to the ablation.

Peri-op Management

Post-op management is super important as it's the difference between a flap living and dying in many scenarios! There's some interesting research highlighted in the syllabus for this. Higher perioperative fluid volume was associated with more postoperative complications after fibular free flap reconstruction, with meaningful cutpoints around 5,500 mL for any complication and 7,000 mL for a major one (Ettinger Microsurgery 2017). Separately, a comparison of ICU versus floor-based postoperative management found no difference in flap failure rate between the two, while the ICU group had a longer length of stay and higher hospital cost in case you were wondering if there was any background research on us sending our patients to Rhoads 3 instead of the SICU (Arshad Head Neck 2014). Neither paper is telling you to abandon careful monitoring, but both are a reminder that more intervention isn't automatically safer, and that fluid stewardship is as much a reconstructive decision as flap choice.

Efficiency matters too, and not just for OR schedules. A large cohort study found that team-based strategies (a standardized surgical team, treatment centralization, and a microsurgery fellowship program) were associated with a roughly 32% reduction in operative time with no increase in complications (Alhefzi JAMA Otolaryngol Head Neck Surg 2023). Shorter operative time is protective in its own right (less time under anesthesia, less time for a flap to sit ischemic), which makes the case that a well-drilled team is itself a complication-prevention strategy.

Plate exposure is usually the more straightforward of the two complications to manage: a small area of exposed hardware without infection can often be handled conservatively with local wound care, while progressive exposure, infection, or plate mobility usually means plate removal, sometimes with a second vascularized flap if the bone stock is compromised (in the setting of ORN of the fibula itself).

Osteoradionecrosis

Osteoradionecrosis deserves its own staging framework rather than a blanket "wound care versus surgery" decision, and it's worth knowing the degrees of disease before deciding how aggressive to be. Mike Fritz's group at Cleveland Clinic has published an updated staging and treatment algorithm that marks a real shift away from older paradigms: hyperbaric oxygen has fallen out of favor as first-line therapy, with growing evidence instead supporting pentoxifylline and tocopherol for early-stage disease (Fritz Otolaryngol Head Neck Surg 2024). For moderate disease, meaning bone that's compromised but hasn't progressed to the point of requiring a segmental resection, their anterolateral thigh fascia lata (ALTFL) rescue flap is the technique worth knowing. Rather than resecting native bone and reconstructing with a vascularized bone flap right away, the fascia lata paddle alone is used to revascularize the remaining mandible in place, and in their original series this arrested disease progression in 22 of 23 patients without ever committing to a segmental reconstruction (Meleca Laryngoscope 2021). I have not had a similar experience with my own patients but I only have a small cohort with this approach thus far. Chronic exposure of the bone without pain or infection can sometimes be monitored without surgical intervention, however progression of this bone exposure often warrants a more aggressive bony resection/reconstruction prior to pathologic fracture. Ultimately when a patient presents with pathologic fracture, unremitting infection, or bone loss too extensive to revascularize in place, a segmental mandibulectomy and osteocutaneous free flap are a more straightforward choice. Lastly, the decision to resect bone but only reconstruct with soft tissue is sometimes the appropriate decision for the right patient. The practical takeaway that runs through this whole email is the same one: the decisions that prevent or limit a complication almost always happen earlier than the complication itself, at flap selection, at the VSP planning session, at the fluid orders you write before the case starts and the post-op management of arrhythmias, low blood pressure, hematoma, or hypercoagulability. These are complex patients who require complex surgical and peri-operative management - thanks for being there as we go through the motions on these cases and for being there for these patients!

Have a great week!

Rob