Experimental swine models for perforator flap dissection in reconstructive microsurgery

Abstract

Background: Perforator flaps account for a fraction of reconstructive procedures despite their growing popularity. Specific microsurgical skills are required for successful harvesting of perforator flaps, which are difficult to attain through direct operating room training. Cadaver and small animal dissection cannot simulate human perforator dissection, lacking either bleeding and vessel feedback or providing too small calibers. Thus, we have developed and refined over the last ten years five perforator flaps models in living pig, described their harvesting technique and provided evidence for their effectiveness as perforator flap training models.

Method: CT angiography data from ten living pigs was used for detailed examination of the integument's vascular anatomy. Microsurgical techniques were used to standardize and harvest the perforator flaps in acute models. The same operator-assistant team, with no prior perforator flap harvesting experience, raised all flaps in a sequential manner, one animal per day, during a 7 weeks timespan. Porcine flaps were compared to human counterparts in terms of vessel caliber, dissection times. Immediate flap survival was measured as duration of perforator pulsation after completion of flap harvesting, measured every 10 minutes for up to two hours.

Results: Five perforator flaps were standardized, based on the deep cranial epigastric, thoracodorsal, lateral intercostal, cranial gluteal and dorsal cervical arteries and the operative technique was described in detail. Mean pig perforator size was 1.24±0.36 mm and mean pedicle diameter was 2.78±0.8 mm, which matched closely the human calibers for each flap. Total harvesting time increased 22.4% between the first two experiments due to a more cautious approach following the lack of perforator pulsation in all flaps in the first experimental animal. A total decrease of 44.4% harvesting time between second and last experiment was observed, as expected with all repetitive surgical procedures. Post-operative perforator pulsation time revealed a steep learning curve, with no or short-term pulsatile perforators in the first five pigs, followed by a 275% increase in total perforator pulsation time between 5th and 6th experimental animal. Based on these findings we provide a description of the most common mistakes, their consequences and gestures which can be trained using the pig perforator flaps, in order to overcome these mistakes.

Conclusion: These five pig perforator flap models provide a fast and efficient learning tool to develop perforator flap harvesting skills safely. Surgical training using these five experimental models offers a similar hands-on perforator flap dissection experience as with human tissue, based on the similar sized calibers of both perforators and pedicles with their human counterparts.

Fig 1. Deep cranial epigastric artery perforator flap planning in pigs.

(A) DCEAPf planning between the 2^nd and 4^th nipple line, centered on the deep cranial epigastric artery (DCEA) and including the medial and lateral row of skin perforators. ITA–internal thoracic artery; SCEV–superficial cranial epigastric vein; DIEA–deep inferior epigastric artery. (B) Dissection of lateral row of perforators. (C) Dissection of medial row perforators. Dotted line marks the incision along the anterior rectus muscle sheet. (D) Epigastric artery dissected and transected distal to the skeletonized perforators, with no sign of vessel spasm. Republished from personal archive under a CC BY license, with permission from Dr. Nistor Alexandru, original copyright 2016.

Fig 2. Thoracodorsal artery perforator flap (TDAPf) and lateral intercostal artery perforator flap (LICAPf) planning in pigs.

(A) TDAPf can be harvested with a dorsal or ventral skin paddle, based on either the horizontal transverse branch or the vertical descending thoracodorsal artery branch. In both cases, the flap extends in the territory irrigated by the lateral intercostal perforators, which pierce the latissimus dorsi on their way to the skin and can be mistaken for the TDA perforator. (B) An extended TDAPf harvested by combining the dorsal and ventral TDAP flap paddles. The latissimus dorsi muscle has been split to allow for a longer pedicle, which can be extended up to the thoracodorsal artery for a total length of 12–14 cm. (C) A large LICAPf can be harvested if no TDAPf is planned on the same hemitorax. Both latissimus dorsi and intercostal muscles are split to allow for a longer pedicle. Republished from personal archive under a CC BY license, with permission from Dr. Nistor Alexandru, original copyright 2016.

Fig 3. Cranial gluteal artery perforator flap (CGAPf) planning in pigs.

(A) Cranial gluteal artery (CGA) perforator pierces first the gluteus medius m. followed by gluteus superficialis m. on its way to the skin and is consistently located 6–10 cm caudal and 3–6 cm ventral from the major trochanter. (B) The CGA terminal branches are the gluteus superficialis m. pedicle and the CGA perforator, which has a long intramuscular course. The gluteus superficialis m. pedicle has to be transected (red dotted line) in order to reach the CGA, about 6 cm deep. Republished from personal archive under a CC BY license, with permission from Dr. Nistor Alexandru, original copyright 2016.

Fig 4. Deep cranial artery perforator flap (DCAPf) planning in pigs.

(A) The DCAPf can be designed either as a half-flap, allowing two surgeons to work simultaneously, or as a single extended flap based on one or multiple perforators, which can be either musculocutanous, septocutaneous or both. The spinous process of C4 and T1 vertebrae serve as bony landmarks for flap planning. (B) Dissecting from the lateral margin of the flap over the brachiocephalic and trapezius m., the septocutaneous perforators are uncovered first, which arise from the DCA. (C) The flap paddle can be vascularized by either muscular or septal DCA perforators, or both. (D) Dissection of the muscular perforator leads to exposure of the deep cranial artery. Republished from personal archive under a CC BY license, with permission from Dr. Nistor Alexandru, original copyright 2016.

Fig 5. Descriptive comparison of perforator and pedicle diameter for the pig models and human counterpart for each flap type.

DCAPf has no human correspondent.

Fig 6. Descriptive overall comparison between the average diameters for all four flap types in pigs and humans.

DCAPf has no human correspondent.

Fig 7. Difficulty of harvesting each flap.

Was assessed based on the mean time required for each perforator flap type to be harvested. ns marks no significant statistical difference; * marks a significant statistical difference, with P<0.05.

Fig 8. Total duration of perforator flap harvesting time.

Was calculated as the sum of all individual perforator flap durations required for each of the five flaps performed in each of the 10 experimental animals. ns marks no significant statistical difference; * marks a significant statistical difference, with P<0.05; ** marks a very significant statistical difference, with P<0.01.

Fig 9. Plotting of the perforator flap learning curve (red) in pig models by aggregating the total duration of observed perforator pulsation after dissection of the perforator flap.

A longer observed pulsation implies a better surgical outcome. ns marks no significant statistical difference; * marks a significant statistical difference, with P<0.05; ** marks a very significant statistical difference, with P<0.01; *** marks a highly significant statistical difference, with P<0.001.