April 2020 HSC Section 4 - Plastic and Reconstructive Problems

Research Original Investigation

Indocyanine Green Angiography and Factors Associated With Perfusion of Paramedian Forehead Flaps

Table 3. Multivariable Analysis

β (95% CI) Outflow Ratio

Factors

Ingress Ratio

Tobacco use

0.075 (−0.133 to 0.283) −0.011 (−0.344 to 0.321) 0.014 (−0.003 to 0.031)

−0.076 (−0.261 to 0.110) 0.026 (−0.269 to 0.320) 0.015 (0.001 to 0.030)

Radiotherapy exposure Time between stages, d

Defect depth

Partial thickness

1 [Reference]

1 [Reference]

Full thickness

0.054 (−0.223 to 0.331) 0.004 (−0.002 to 0.011) −0.240 (−0.472 to −0.008)

−0.106 (−0.352 to 0.140 0.003 (−0.003 to 0.008) 0.039 (−0.169 to 0.247)

Defect area, cm 2 Cartilage graft use

with time, whereas the flapwas both perfusedwith blood and drained at a slower rate than the cheek. As shown by the low rate of complications, these mean perfusion values were suf- ficient to provide adequate blood supply to the flap after di- vision of the pedicle. In this study, nomajor complications and a 3% rate of minor complications were reported, which re- flects a relatively low complications rate of this surgical pro- cedure compared with reported complications rates varying between 6.1% and 14.1%. 19,20 Use of ICG angiography re- sulted inno instances of complete flapnecrosis,whichhas been reported in other studies at rates between 1.2% and 2.0%. 19,21 In this study, patient factors investigated includedage, sex, tobacco use, diabetes, and radiotherapy exposure. Other than diabetes, none of these factors showed consistent associa- tions with the neovascularization perfusion measures as- sessed. Diabetes had a negative associationwith all perfusion measures; however, this finding was not statistically signifi- cant. This observation was supported by previously pub- lished studies that used color duplex–Doppler ultrasonogra- phy to assess flap perfusion and reported no association between age, sex, or tobacco use and the resistance index of blood flow in forehead flaps. 7 Although tobaccousewas shown to be associated with greater odds of flap necrosis in some studies, 21 others showed that it did not have a significant as- sociation with wound dehiscence, scars, flap necrosis, or notching. 19,22 Factors of the recipient site (defect) evaluated included pathologic diagnosis, defect surface area, defect depth (par- tial vs full thickness), cartilage graft use, and time between stages. As expected, longer time between the 2 stageswas posi- tively associatedwith the flap-to-cheek ingress ratio, suggest- ing that arterial inflow to the flapwas faster with increased du- ration of time between stages. A negative association was observed between the flap-to-cheek outflow ratio and carti- lage graft use, suggesting that cartilage graft usemay slowve- nous neovascularization of the flap. One implication of this finding was that venous outflow may be more dependent on the wound bed than arterial inflow, although this was not in- dependently examined in this study. Examination of other re- cipient site factors’ association with perfusion measures did not reach statistical significance. In previous studies, full- thickness defects were associated with significantly higher rates of developing a major complication, including flap ne- crosis, notching, and obstruction, compared with partial- thickness defects. 21 Although no statistically significant

Factors associated with greater time between stages by linear regression included tobacco use (β, 3.93; 95% CI, 0.76-7.10), radiotherapy exposure (β, 5.05; 95% CI, 0.01-10.10), full- thickness defect (β, 6.76; 95% CI, 3.18-10.33), defect surface area (β, 0.15; 95%CI, 0.06-0.24), andcartilage graft use (β, 5.28; 95% CI, 2.17-8.39). These factors, along with time between stages, wereused in themultivariable linear regression for flap- to-cheek outflow ratio and flap-to-cheek ingress ratio to iden- tify factors independently associated with flap perfusion. The multivariable linear regression of flap-to-cheek out- flow ratio and flap-to-cheek ingress ratio is shown in Table 3 . Regarding the flap-to-cheek outflow, cartilage graft usewas re- ported tohave statistically significant negative associationwith flapperfusion (β, –0.240; 95%CI, –0.472 to –0.008). Other fac- tors did not reach statistical significance on multivariable re- gression; however, time between stages, which was statisti- cally significant in the univariable model, showed the next closest association with flap perfusion (β, 0.014; 95% CI, –0.0002 to 0.030). When assessing flap-to-cheek ingress ra- tio, time between stages had a statistically significant inde- pendent associationwith flapperfusion (β, 0.015; 95%CI, 0.001 to 0.030). Discussion Although forehead flaps are themost commonly used type of flap to reconstruct large nasal defects, there are few studies assessing flap perfusion and hemodynamics and their clini- cal implications. To our knowledge, this was the largest study of patients undergoing intraoperative perfusion assessment of forehead flaps used for nasal reconstruction. This study pro- vided quantitative relative perfusion data and clinical out- comes of the 2-stage nasal reconstruction. As shown inTable 2, the time-independent perfusionmeasures assessing flow (flap ingress [arterial inflow] and flap egress [venous outflow]) were both less than 1, signifying slowing perfusion of the flap com- pared with the cheek. This finding was also supported by the time-dependent factors assessing fluorescence of the flap com- pared with the cheek (flap-to-cheek inflow ratio, peak fluo- rescence ratio, outflow ratio, and the calculatedmean of these measures). Allmean results reportedwere less than 1, and they increasedwith time (0.48 [inflow ratio] to 0.59 [peak fluores- cence ratio] to 0.88 [outflow ratio]), meaning that the flap be- came increasinglymore fluorescent comparedwith the cheek

JAMA Facial Plastic Surgery May/June 2019 Volume 21, Number 3 (Reprinted)

jamafacialplasticsurgery.com

© 2019 American Medical Association. All rights reserved.

140