In the 1970s, Behan and Wilson1 introduced the angiotome concept that culminated in the keystone island flap design.2,3 An emerging body of basic science research and published clinical series suggest that keystone island flaps are robust and versatile with superior clinical outcomes due at least partially to its island design.4–6 Based on his clinical observations including the “red dot sign” and hyperemic flare,7 Behan has advocated the superior vascularity of the island flap design for at least 2 decades. Despite increasing interest and popularity of the keystone island flap4,8,9 in the current era of evidence-based medicine, his clinical observations and interpretations need to be further substantiated.
The aim of this study is to determine (1) whether surgical islanding of a flap alters the vascularity or blood supply of the flap and (2) whether these changes in blood supply explain Behan’s clinical observations of “red dot sign” and hyperemic flare.
This study was approved by Human Research and Ethics Committee on September 6, 2013 (reference: 131 93B). Subjects for this study were recruited by the first author from patients undergoing local island fasciocutaneous flap or anterolateral thigh fasciocutaneous free flap reconstructions over a 10-month period from September 2013 to July 2014 at a single institution. Every subject provided oral and written consent via a Participant Information Sheet and Consent Form approved by the Human Research and Ethics Committee. Data collected include patient demographics, clinical details, outcomes including complications, and follow-up period.
Collection of Histological Specimens
Each specimen consisted of 10 mm3 blocks of soft tissue (skin and subcutaneous tissue) removed from the patient at specified times. After a traumatic harvest, each specimen was placed in containers with 4% buffered formaldehyde. Vasoconstrictive or vasodilating agents, such as local anesthesia and adrenaline, were not used, and diathermy and electrocautery were avoided to minimize tissue damage in the specimens.
After flap design and skin markings made as per normal routine, 3 adjacent specimens were collected from each patient during their surgery. First, a control specimen was taken (Fig. 1). Then a flap was raised to simulate a transposition flap with at least 25% of the skin bridge remaining intact. The second specimen was then harvested from the tip of the transposition flap (non-island flap specimen). The remaining skin bridge was then divided and the islanding completed. Last, the third specimen (island flap specimen) was harvested from the island flap, from an area adjacent to the first 2 specimens.
For each patient, 3 specimen containers were labeled with a square, triangle, and circle, respectively, and were placed in a plastic bag. As each specimen (control, non-island, or island) was harvested, a theater nurse chose a labeled container at random to receive the specimen, with the final specimen going into the remaining container in the plastic bag. As records of the specimen and corresponding containers were made available only to the first author, the pathologist assessing the vasculature of the specimens was completely masked to the specimen type.
For each specimen, multiple sections were performed (minimum of 2 sections). After routine hematoxylin and eosin (H&E) staining, the histopathologist examined arteriolar wall thickness, intraluminal arteriolar diameter, venule wall thickness, and intraluminal venule diameter (in μM using an eyepiece reticule at 400× total magnification). Ten separate fields were assessed for each of these variables; consequently each specimen, for each patient, had 10 measurements for each of the 4 microvascular variables being studied.
Data were available from 13 patients, and the data from all patients were aggregated according to the type of specimen (control, non-island, and island), and the 4 microvascular variables being assessed. The raw data were retained, and in addition, we created a parallel database with log-transformed data as the distribution of the raw data was not normal.
Statistical analyses and graphing were done using Prism version 6.0 for Mac OS X. Categorical data were analyzed by the χ2 test. Nonparametric analog data from all 3 stages were assessed using the Kruskal–Wallis nonparametric test, and the log-transformed (normalized) data were tested by one-way ANOVA. Secondary comparisons between 2 of the 3 specimens used the Mann–Whitney test for nonparametric data and t-tests for normalized data. All comparisons used P < 0.05 as indicating significance.
RESULTS AND DISCUSSION
To our knowledge, this is the first histological study comparing non-island and island flaps. Thirteen patients (with 14 sets of specimens) were initially recruited for this study (Table 1). One patient was excluded from further analysis due to inadvertent departure from protocol during collection of the specimens. Twelve patients (with 13 sets of specimens) were analyzed histologically (Table 2).
Overall, there were significant changes in arteriole diameter and venule diameter as the plastic surgical procedure progressed from control to non-island and then on to island (Fig. 2). The big change in arteriole diameter was an increase from control to the non-island state, with a nonsignificant (P = 0.18) decrease from non-island to island. Venule diameter increased gradually and reached a plateau during progression from non-island to island flaps.
Interestingly, the change in arteriolar wall thickness was only a trend, and the change in venular wall thickness was not significant. Arteriolar wall thickness increased from control to the non-island state but then regressed to the control value in the island specimen.
These results were consistent with Behan’s clinical observations of “red dot sign” and hyperemic flare of more than 2 decades. He noted that island flaps were relatively pink or vascular in complexion (hyperemic flare). While insetting island flaps, they almost always bled from the site where the suture needle pierced the flap, an observation not replicated by the opposite skin edge (red dot sign). The histological findings of the increased arteriole and venule diameter in the island flaps support these observations of increased vascularity in the island flap.
The high venule diameter in the non-island and island states (relative to the control) is likely associated with increased venous perfusion pressure and some degree of venous congestion. However, islanding led to decreased arteriole diameter and presumably decreased inflow, potentially helping to ease congestion and improve flap survival.
Retention of a dermal pedicle, as is the case in the design of a transposition flap as opposed to an island flap, may alleviate anxiety. However, it has been previously demonstrated that island flaps survive to at least the same length as those with a cutaneous pedicle, which contains segmental vessels.10 More recently, it was shown that conversion of a perforator flap with a skin bridge into an island perforator flap prevented “hemodynamic steal” and increased peripheral tissue perfusion.11 This may form the basis for improved survival of island flaps relative to dermal pedicled flaps. Based on studies and clinical impressions past and present (ours), retention of a skin bridge brings no added advantage.10,12 In addition, we concur with the suggestion of incorporating named or segmental underlying vessels, and hence, our principle of designing keystone flaps was based on the angiotome concept.7
The sequence of vasoconstriction and the coagulation cascade following trauma may be well documented, but the control of blood supply to the skin is yet to be fully understood. Cutaneous blood flow is the result of a complex interplay of reflex (whole body) and local control mechanisms.13 Reflex sympathetic innervation of cutaneous circulation has 2 branches; sympathetic noradrenergic vasoconstrictor system and non-noradrenergic active vasodilator system.13,14 Sympathetic noradrenergic vasoconstrictor nerves provide tonic innervation. Interruption of this sympathetic noradrenergic innervation causes a passive vasodilatation due to withdrawal of the tonic activity of vasoconstrictor nerves.13 The active vasodilator system does not exhibit resting tone and is only activated by increases in body temperature (heat exposure and exercise).13 A role also exists for afferent or sensory nerves.15
Observations similar to Figure 2 were made when data from the subgroups of local island flaps and anterolateral free flaps were analyzed separately. These findings confirm our assumption that the vascular changes occurring in both subgroups occur along the same spectrum with the raising of these flaps.
This study involves histological assessment of specimens harvested in the early intraoperative period. Although the small number of patients could be considered a limitation of this study, the highly significant differences suggest that a much larger study would be unlikely to arrive to a different outcome. To investigate intraflap differences, specimens need to be harvested from the base and the tip of the transposition flap. To verify permanence of these vascular changes, and ultimately impact on flap survival, a long-term study is necessary. The effect of flap mobility, flap insetting, and tension on flaps with altered vascularity remains outside the scope of this study.
The histological findings of increased arteriole and venule diameter in the island flap were entirely consistent with Behan’s clinical observations of red dot sign and hyperemic flare. Further studies are required to directly compare island and non-island flap designs.
We thank surgical and nursing colleagues at Monash Health who assisted in the collection of histological specimens.
1. Behan FC, Wilson JSPPresented at: The Second Congress of the European Section of the International Confederation of Plastic and Reconstructive Surgery, May 1973, Madrid, Spain. . The vascular basis of laterally based forehead island flaps, and their clinical applications.
2. Behan FC, Wilson JSPHueston J. The principle of the angiotome, a system of linked axial pattern flaps.
Transactions of the Sixth International Congress of Plastic and Reconstructive Surgery. 1975 Paris;
3. Behan FC. The keystone design perforator island flap in reconstructive surgery. ANZ J Surg. 2003;73:112–120
4. Pelissier P, Santoul M, Pinsolle V, et al. The keystone design perforator island flap. Part I: anatomic study. J Plast Reconstr Aesthet Surg. 2007;60:883–887
5. Behan F, Sizeland A, Gilmour F, et al. Use of the keystone island flap for advanced head and neck cancer in the elderly─a principle of amelioration. J Plast Reconstr Aesthet Surg. 2010;63:739–745
6. Behan FC, Lo CH, Sizeland A, et al. Keystone island flap reconstruction of parotid defects. Plast Reconstr Surg. 2012;130:36e–41e
7. Behan FC, Findlay M, Lo CH Concept & Applications: Keystone Island Perforator Flap. 2012 Australia Elsevier;
8. Shipkov CD, Mojallal A. The Keystone island and pedicle flap: a handy local flap for soft tissue reconstruction. Ann Surg Oncol. 2008;15:3625
9. Jackson IT. The keystone design perforator island flap in reconstructive surgery. ANZ J Surg. 2003;73:261
10. Milton SH. Experimental studies on island flaps. 1. The surviving length. Plast Reconstr Surg. 1971;48:574–578
11. Mešić H, Kirkebøen KA, Bains R. The importance of a skin bridge in peripheral tissue perfusion in perforator flaps. Plast Reconstr Surg. 2012;129:428e–434e
12. Moncrieff MD, Bowen F, Thompson JF, et al. Keystone flap reconstruction of primary melanoma excision defects of the leg-the end of the skin graft? Ann Surg Oncol. 2008;15:2867–2873
13. Charkoudian N. Mechanisms and modifiers of reflex induced cutaneous vasodilation and vasoconstriction in humans. J Appl Physiol (1985). 2010;109:1221–1228
14. Hodges GJ, Johnson JM. Adrenergic control of the human cutaneous circulation. Appl Physiol Nutr Metab. 2009;34:829–839
© 2016 American Society of Plastic Surgeons
15. Johnson JM, Kellogg DL Jr. Thermoregulatory and thermal control in the human cutaneous circulation. Front Biosci (Schol Ed). 2010;2:825–853