April 2020 HSC Section 4 - Plastic and Reconstructive Problems

Original Investigation Research

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

Table 1. Baseline Patient and Recipient Site Characteristics Characteristic (N = 71)

Table 2. Flap-to-Cheek Perfusion Measures Flap-to-Cheek Perfusion Measure Time-dependent ratio Midpoint of indocyanine green flap inflow

Patients a

Value, Mean (SD)

Sex

Male

43 (61) 28 (39)

0.48 (0.40) 0.59 (0.34) 0.88 (0.42) 0.65 (0.35)

Female

Peak fluorescence

Age, mean (SD), y Pathologic diagnosis

71.1 (11.0)

Midpoint of indocyanine green outflow Calculated mean of inflow, peak, and outflow measures

Basal cell carcinoma

54 (76)

Time-independent ratio Flap ingress or arterial inflow Flap egress or venous outflow

Squamous cell carcinoma

8 (11)

0.54 (0.36) 0.65 (0.98)

Melanoma

6 (8) 3 (4)

Other b

Tobacco use

31 (44)

The assessed patients did not experience flap necrosis or nasal obstruction. However, 2 patients (3%) experienced mi- nor complications: 1 patient had distal superficial epidermal sloughing, and another patient had partial distal wound de- hiscence. One patient was a nonsmoking woman in her 80s with a history of deep vein thrombosis and a 13-day interval between the 2 procedure stages. The other patient was a man in his 60s with a history of tobacco use who had a full- thickness defect and cartilage graft with a 23-day interval be- tween procedure stages. Perfusion data obtained by each method are shown in Table 2 . The mean (SD) flap-to-cheek inflow ratio was 0.48 (0.40), the peak fluorescence ratio was 0.59 (0.34), and the outflow ratio was 0.88 (0.42). The calculated mean (SD) flap- to-cheek perfusion ratio of these measures (inflow, peak fluorescence, and outflow) was 0.65 (0.35). The mean (SD) flap-to-cheek ingress ratio was 0.54 (0.36) and egress ratio was 0.65 (0.98). All of these measures showed normal distri- bution except the flap-to-cheek egress ratio, which showed greater variability in values. These data revealed that there was a greater relative amount of fluorescence during blood outflow from the flap compared with during inflow, suggest- ing that the flap filled and drained at a slower rate than the cheek; similarly, the mean flap-to-cheek ingress ratio was lower than the egress ratio, signifying more venous conges- tion of the flap compared with the cheek. Each perfusion measure was analyzed by univariable lin- ear regression; the patient and defect characteristics are listed in the eTable in the Supplement . Only 2 perfusionmeasures— the flap-to-cheek outflow ratio and the flap-to-cheek ingress ratio—were statistically significantly associated with indi- vidual patient or defect characteristics, and thus only these 2 outcomevariableswereused to create themultivariablemodel. Although not statistically significant, diabetes had a consis- tently negative associationwith flap perfusion across all 6 per- fusionmeasures (β ranging from−0.38 to −0.107), whereas to- bacco use, radiotherapy exposure, and defect characteristics hadmorevariable associations (β ranging from−0.242 to0.256) (eTable in the Supplement ). The time between flap placement and pedicle division and flap inset had the greatest association with perfusion mea- sures using the univariable linear model. Further assessment was performed regarding the association between this vari- able andother patient or defect characteristics to assess for col- linearity and aid in construction of the multivariable model.

Head and neck radiotherapy exposure c

8 (11)

Diabetes

12 (17)

Time between stages, mean (SD), d d

21.3 (6.9)

Defect depth

Partial thickness

55 (77) 16 (23)

Full thickness

Defect surface area, mean (SD), cm 2

14.2 (17.3)

able linear regression model. A P < .05 was considered to be statistically significant; β values of the linear regressions and 95% CIs were reported. b Other includes defects caused by infection or trauma. c Five of 8 radiotherapy exposures were directly to the nose. d Stage 1 defined as flap placement and stage 2 as pedicle division with flap inset. Results Of 85 patients who met the inclusion criteria for this study, complete angiography data were available for 71 patients; 14 patients were excluded because of absent or corrupted angi- ography data files owing to patientmovement during data col- lection. Baseline patient and defect characteristics are shown in Table 1 . Most of the patients in this study were male (43 of 71 [61%]), with a mean (SD) age of 71.1 (11.0) years. Most pa- tients had a diagnosis of basal cell carcinoma (54 [76%]), fol- lowed by squamous cell carcinoma (8 [11%]), melanoma (6 [8%]), and reconstruction because of other defects including infection or trauma (3 [4%]). Comorbidities assessed in- cluded tobacco use (31 [44%]), radiotherapy exposure of the head and neck (8 [11%]) with 5 exposures directly to the nose, and diabetes (12 [17%]). The mean (SD) interval between flap placement and pedicle division with flap inset was 21.3 (6.9) days. Defect depth included partial thickness defects in 55 pa- tients (77%), full thickness defects in 16 patients (23%), mean (SD) defect surface area of 14.2 (17.3) cm 2 , and cartilage graft use in 43 patients (61%). The excluded 14 patients did not have characteristics significantly different from the study popula- tion after χ 2 or independent t test analysis except less carti- lage graft use (61%vs 28%; P = .02), and they did not encoun- ter more complications. Cartilage graft use 43 (61) a Data are presented as number (percentage) of patients unless otherwise indicated.

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

jamafacialplasticsurgery.com

© 2019 American Medical Association. All rights reserved.

139