Robotic plastic surgery: past, present and future

Breast reconstructive surgery: applications in incisionless harvest of the latissimus dorsi flap

Since its earliest description in 1906³⁰, the latissimus dorsi (LD) muscle flap has been an essential workhorse for reconstructive surgery; it has since gained great popularity in breast reconstruction, whether in second stage (after expansion) or in primary reconstruction. This flap is typically harvested with a skin paddle providing an intact non-irradiated skin to replace the irradiated area. Raising the muscle alone (without skin) has also a number of applications in breast reconstruction, including:

• Protection of implants in stage 2 breast reconstruction, following expansion
• First stage of an immediate or two-stage implant-based reconstruction following Nipple-Areolar Complex (NAC) sparing mastectomy³¹
• Partial breast reconstruction following partial mastectomy of the outer quadrants
• As an alternative to bioprosthetic mesh to support the lower pole of implants, obviating the need for biologic materials and their additional cost and infection risks^31, 32.

Harvesting the latissimus dorsi flap requires a long incision on the back that ranges between 15 and 45 cm, in addition to an axillary incision for pedicle isolation and transfer^33, 34.

When the skin component of the flap is not required for reconstruction (as in the aforementioned scenarios), a minimally-invasive harvest technique with no scars on the back would be very favourable with regard to donor site morbidity. This has been attempted using lighted retractors and long instruments, known as the semi-open approach, where the incision size was reduced to 5–8 cm (by 80–88%). However, this approach is technically challenging, and surgeons commonly have to extend their incision to complete their dissection at the paraspinal level. Endoscopic harvest has also been attempted, but because of technical challenges related to limitations of endoscopic instrumentation and difficulties in maintaining an adequate optical window — especially when dissecting around the curvature of the back — nearly all centres have abandoned this technique^35–38. The robotic platform, on the other hand, confers unique benefits that overcome the limitations of the traditional and endoscopic approaches, specifically:

• High resolution with great picture clarity and magnification, allowing better identification and control of perforators (a common problem in the endoscopic approach)
• Seven degrees of freedom of robotic instruments allowing incredible precision, maintaining a consistent plane, and offering better negotiation around the curvature of the back
• Surgeon comfort and ease of dissection, limiting any mechanical disadvantage when meticulous dissection is performed
• Needless to say, aesthetics and donor site cosmesis, especially for breast reconstruction patients, are significant determinants of final outcomes; robotic incisions of the muscle-only LD flap are barely visible³⁹.

First introduced by Selber in a cadaver model, then in a clinical series of eight patients, robotic harvest of the LD muscle-only flap is a feasible technique that allows effective coverage and provides a reliable reconstruction with well-concealed incisions. In addition, it is safe, with decreased donor site morbidity and no major complications^31, 40. The use of this technique has been steadily increasing in recent years; Selber performs an average of 15 robotic LD harvests for breast reconstruction per year. The main indications for the robotic technique include reconstructions of lateral defects post-partial mastectomy, implant-based reconstruction post-NAC sparing mastectomies, and in patients with expanders who receive radiation. Owing to the safety and tremendous advantages of this technique, its indications are increasing³⁹.

The average set up time (including initial axillary incisions, port placement, and docking of the robot) is approximately 30 minutes. The actual harvest itself takes a little over 1 hour.

No conversion to open technique has occurred so far, and all flaps were harvested and transferred in their entireness. Most importantly, no surgically-related complications such as haematoma, seroma, or overlying skin injury have occurred³⁹.

It is worth noting that this technique has a certain learning curve for it to be performed in a safe and efficient manner. Studies are needed to assess the steepness of such a learning curve, but it is recognised that robotic‑assisted procedures are in general much easier to learn than laparoscopic procedures.

Abdomino-pelvic reconstructive surgery: applications in morbidity-free rectus muscle and omental flap harvest

Pelvic defects after oncological resections (being colorectal, gynaecological or urological) are common and challenging reconstructive problems for plastic surgeons^41–44. These types of resections are extensive and leave a substantial dead space that, if not replaced with vascularised tissue, would lead to high rates of infection and abscess formation⁴⁵. The pedicled rectus abdominis muscle flap is a first-choice reconstruction for such defects. In cases where the defect is larger than usual and a muscle flap alone would not be enough for adequate coverage, a pedicled omentum or a de-epithelialised myocutaneous vertical rectus abdominis myocutaneous (VRAM) can be added⁴⁶. In cases where skin or vaginal wall are missing and a replacement is needed, a VRAM flap is then an essential requirement.

Traditionally, the rectus muscle is accessed through the same abdominal laparotomy incision that was created by the oncological surgeon; this is followed by vertical division of the anterior rectus sheath along the length of the entire muscle. This approach carries substantial postoperative complications (such as wound infection, seroma, abdominal hernia and pain), and has undesirable cosmetic results. Reported rates of bulge and hernia range from 1.7–4% and 0.85–2.9%, respectively^47–51. Overall, surgical site morbidity incidence ranges from 8.5 to 14.3%^{47, 48, 52}.

The concept of a minimally-invasive approach that does not entail violating the anterior rectus sheath to access the rectus muscle is very appealing for reconstructive surgeons. This was introduced by endoscopic extraperitoneal and laparoscopic transperitoneal approaches to harvest the rectus muscle^53–55. These techniques did not gain popularity owing to their limited success; the endoscopic technique still violated the anterior rectus sheath to varying degrees, and the laparoscopic technique required advanced skills to be performed⁵⁶.

Robotic surgery, on the other hand, has gained wide applications in pelvic oncologic procedures. Manipulation of the robot is more intuitive than laparoscopy and is much easier to operate as proven by a lower learning curve^57, 58. This offers significant advantages for plastic surgeons (and their patients alike) for performing minimally-invasive procedures, such as robotic harvest of the rectus abdominis muscle. Whenever needed, the omentum can be harvested in a similar fashion.

Intraperitoneal robotic-assisted harvest of the rectus muscle was first developed on a cadaver model by Selber and Pederson in 2009⁵⁹. It was translated clinically by Pederson in 2010, and since then many cases have been performed at different institutions⁶⁰. The salient lessons learned from this procedure so far, are:

• Only three trocars are needed in the contralateral quadrants of the hemi-abdomen. If needed, a fourth trocar can be inserted, free-style, to help the first assistant in difficult situations
• The presence of the plastic surgeon during trocars insertion is essential
• The highlight of this approach is avoidance of any violation of the rectus muscle and its anterior sheath.

Experience with the robotic approach has so far proven to be superior to the laparoscopic method, as well as to the endoscopic-assisted technique for harvesting the rectus muscle. This is mainly owing to the ease and simplicity of dissection (as compared with laparoscopy, in which it is almost impossible for the human arms to direct the instruments towards the midline when performing the initial dissection), and, as outlined above, the preservation of the anterior rectus sheath with a potential decrease in the incidence of hernia (as compared to the endoscopic approach).