Risk Factors for Partial Flap Loss in a Free Flap: A 12-Year Retrospective Study of Anterolateral Thigh Free Flaps in 303 Lower Extremity Cases : Plastic and Reconstructive Surgery

Journal Logo

Reconstructive: Lower Extremity: Original Articles

Risk Factors for Partial Flap Loss in a Free Flap: A 12-Year Retrospective Study of Anterolateral Thigh Free Flaps in 303 Lower Extremity Cases

Min, Kyunghyun M.D.; Hong, Joon Pio M.D., Ph.D., M.M.M.; Suh, Hyunsuk Peter M.D., Ph.D.

Author Information
Plastic and Reconstructive Surgery: November 2022 - Volume 150 - Issue 5 - p 1071e-1081e
doi: 10.1097/PRS.0000000000009646
  • Free
  • Journal Club
  • Patient Safety CME
  • Podcast

Abstract

In 1989, Koshima et al. introduced the first perforator-based flap, which was an inferior epigastric artery flap containing one perforator without rectus abdominis muscle.1 In addition, since the introduction of the concept of the angiosome presented by Taylor et al., many perforator flaps have been introduced and widely used.2–6 Nowadays, by advances through constant research and development of the techniques and equipment, perforator flaps including perforator-to-perforator microanastomosis and nanomicrosurgery have become common procedures.7–9 Moreover, the perforasome theory by Michel Saint-Cyr et al. has deepened the understanding of flap microcirculation and allowed surgeons to elevate flaps more safely.10

Despite these remarkable developments, total failure of the flap still occurs, and partial loss occurs more commonly than total necrosis. Total necrosis is known to occur because of microanastomosis failure, which is induced by compression of the pedicle, venous hypertension, and flap elevation error; however, the cause of partial necrosis and tips for minimizing partial necrosis of perforator flaps has not been clearly discussed.11,12 Partial necrosis of the flap requires additional time and treatment to repair the necrotized area. Furthermore, delayed wound healing following flap coverage might postpone adjuvant chemotherapy or radiotherapy in cancer patients.

Among many perforator flaps, the anterolateral thigh free flap has been one of the most commonly used since it was first introduced by Song et al. in 1984.13 Surgical anatomy has been well described by several studies regarding angiosome territory.3,14 However, even 4 to 43 percent of anterolateral thigh free flap cases still experience partial flap necrosis.11,15,16 This study aimed to analyze partial necrosis cases that underwent reconstruction of the lower extremities using anterolateral thigh flaps and identify risk factors causing the occurrence of partial necrosis.

Patients and Methods

Patient Selection and Demographics

From January of 2005 to February of 2017, a retrospective study was conducted on cases in which a lower leg was reconstructed using an anterolateral thigh free flap in a single center after approval by the institutional review board (number 2020-1140). We investigated the causes of soft-tissue defects on the lower leg and demographic data such as hypertension, diabetes, chronic kidney disease, and smoking history. In addition, we also evaluated the patients’ physical status according to the American Society of Anesthesiologists score. Patients who had total flap failure, ischemic necrosis, or intraflap microanastomosis (turbocharged or supercharged) were excluded from the study. Also, diabetic foot and ischemic necrosis cases were excluded to analyze partial flap loss with relatively intact vessel status, similar blood pressure, and similar serum glucose levels.

Because there is no consensus on the definition of partial necrosis, we defined partial loss (or necrosis) in this study when full-thickness skin necrosis occurs with a width of at least 1 cm at the margin of the flap and more than 50 percent of the flap is viable at the same time. The partial loss was confirmed 3 weeks after the reconstructive surgery, which is the time when demarcation was completed.

Patient Intraoperative Data and Laboratory Findings

As the intraoperative data, reconstruction site were classified into proximal one-third, middle one-third, distal one-third, and foot. Data regarding the size of the flap and layers of flap elevation were also collected, and the diameter of the harvested pedicle, the number of perforators harvested, the method of arterial anastomosis (end-to-end or end-to-side), and the number of vein anastomoses were investigated. The statistical difference of intraoperative data was analyzed between the partial loss group and the non–partial loss group. In addition, comparison of mean flap size difference by the number of perforators and occurrence of partial loss was also conducted.

To verify whether laboratory findings affect the occurrence of flap partial necrosis, patients’ preoperative and postoperative C-reactive protein, albumin, and glucose level were measured within 2 days before surgery and 5 days after surgery to check the patient’s progress after reconstructive surgery, and the mean blood pressure levels for 5 days after surgery were also evaluated. In addition, the presence of bacteria was assessed.

Partial Loss Analysis Strategies

The following method was used to examine the rate of partial flap necrosis according to the distance between perforator and flap tissue. First, the longest distance from the perforator to the flap margin and the shortest distance from the perforator to the partial necrosis margin were measured using ImageJ software (National Institutes of Health, Bethesda, Md.). If the number of harvesting perforators is more than two, the distance from each perforator to the flap margin was measured after dividing the flap area vertically based on the midpoints of adjacent perforators. Then, the longest distance from the perforator to the designed flap is divided into 1-cm sections, and the number of harvested flaps on each section was collected to act as a denominator. Alternatively, the number of cases of partial necrosis on each section was collected to act as a numerator. Based on those data, the partial loss rate was calculated on each section (Figs. 1 and 2).

F1
Fig. 1.:
The following method was used to examine rate of partial flap necrosis according to the distance between the perforator and flap tissue. First, the longest distance from the perforator to the flap margin and the shortest distance from the perforator to the partial necrosis margin were measured using ImageJ software. Then, the distance from the perforator to the designed flap is divided into 1-cm sections, and the number of harvested flaps on each section was collected to act as a denominator. Alternatively, the number of cases of partial necrosis on each section was collected to act as a numerator. Based on those data, the partial loss rate was calculated on each section.
F2
Fig. 2.:
A case of partial necrosis. (Above, left) A 73-year-old woman visited the clinic for an infection and soft-tissue defect that occurred after bimalleolar fracture of ankle. She had diagnosed well-controlled type 2 diabetes mellitus and hypertension. (Above, right) After serial débridement, a 22 × 8-cm anterolateral thigh perforator free flap was elevated for the reconstruction. The elevation plane was the supradeep fat layer. Two perforators were harvested, and the longest distance from the perforator to the distal flap margin was 11.5 cm. (Below, left) A photograph obtained on postoperative day 10. Partial necrosis occurred on the distal part of the flap. (Below, right) Débridement was performed. The shortest distance from the perforator to the partial necrosis margin was 4 cm. Additional skin graft was required for the defect area.

Statistical Analysis

To compare the difference between the groups with and without partial flap losses, the t test or Mann-Whitney U test was used for continuous variables, and the chi-square test or Fisher exact test was used for categorical variables. If more than three continuous variables were compared, the Kruskal-Wallis test was conducted. In addition, post hoc analysis was performed to check the significance between each variable, and the significant level was corrected using the Bonferroni method. Values of p < 0.05 were used to denote statistical significance in all tests except for post hoc analysis, which were considered statistically significant when the p value was less than 0.0083. Furthermore, the cutoff value of the distance from the perforator to the margin of partial necrosis was measured using a receiver operating characteristic curve to estimate the point that distinguishes the probability of partial necrosis. The cutoff value of the size of the flap was also calculated with the receiver operating characteristic curve, and the area under the curve was calculated in both variables to verify the reliability of the cutoff value. Finally, independent factors increasing the occurrence of partial flap losses were identified through multivariable logistic regression analysis with the backward elimination method. IBM SPSS Version 25.0 (IBM Corp., Armonk, N.Y.) was used to confirm the statistical significance of the data.

Results

A total of 303 cases of lower extremity reconstruction using anterolateral thigh free flaps were included in the study. Acute trauma was the most common cause of defect (128 cases), followed by tumor resection (76 cases), chronic osteomyelitis (64 cases), and unstable wound (35 cases). Of all defects, 11 defects were covered using two anterolateral thigh free flaps simultaneously (acute trauma, nine cases; unstable wound, two cases). In addition, a secondary free flap was required in one case because of tumor recurrence. Finally, two cases underwent secondary anterolateral thigh flap surgery to cover a wide area of partial necrosis of previous flaps.

Comparison of Patient Demographics

Partial flap losses occurred in 43 of the 303 patients included in this study. Between the partial loss group and the non–partial loss group, no variables showed statistical differences in demographic data (Table 1).

Table 1. - Preoperative Demographics*
Characteristic Overall (%) No Partial Loss (%) Partial Loss (%) p
No. of patients 303 260 43
Mean age ± SD, yr 46.21 ± 18.45 46.50 ± 18.31 44.49 ± 19.44 0.5087
Sex 0.8134
 Male 221 (72.94) 189 (72.69) 32 (74.42)
 Female 82 (27.06) 71 (27.31) 11 (25.58)
Hypertension 57 (18.81) 47 (18.08) 10 (23.26) 0.4208
Diabetes 30 (9.9) 25 (9.62) 5 (11.63) 0.5929
History of smoking 127 (41.91) 108 (41.54) 19 (44.19) 0.7445
Chronic kidney disease 1 (0.33) 0 (0) 1 (2.33) 0.1419
History of PTA 15 (4.95) 12 (4.62) 3 (6.98) 0.4551
ASA score 0.1871
 1 64 (21.12) 58 (22.31) 6 (13.95)
 2 229 (75.58) 195 (75) 34 (79.07)
 3 10 (3.3) 7 (2.69) 3 (6.98)
PTA, percutaneous transluminal angioplasty; ASA, American Society of Anesthesiologists.
*The t test was used for continuous variables and the χ2 test or Fisher exact test was used for categorical variables.
Fisher exact test.

Comparison of Intraoperative Factors

As a result of investigating intraoperative factors between the two groups, mean flap size of the partial loss group was 227.40 ± 143.80 cm2, which was wider than that of the non–partial loss group (164.9 ± 102.4 cm2), and it was statistically significant (p = 0.0011). In addition, the rate of partial necrosis was higher, as the layer of flap elevation was thinner and was statistically different (p = 0.0001). In the case of number of vein anastomoses, partial necrosis occurred at a rate of 19.73 percent in cases in which one-vein anastomosis was performed (22 of 113 cases), whereas a 11.05 percent rate of partial necrosis was observed in cases of more than two-vein microanastomoses (21 of 190 cases), which was statistically significant (p = 0.0423). The rate of partial necrosis according to the defect location, number of harvesting perforators, pedicle diameter, and type of arterial anastomosis was not observed (Table 2).

Table 2. - Intraoperative Factors*
Characteristic Overall (%) No Partial Loss (%) Partial Loss (%) p
Mean flap size ± SD, cm2 173.76 ± 111.09 164.90 ± 102.40 227.40 ± 143.80 0.0011
Elevation layer 0.0001
 Subfascial 180 (59.41) 166 (92.22) 14 (7.78)
 Suprafascial 29 (9.57) 24 (82.76) 5 (17.24)
 Supradeep fat 94 (31.02) 70 (74.47) 24 (25.53)
Culture results 0.102
 Negative 53 (17.49) 42 (79.25) 11 (20.75)
 Antibiotic-susceptible 72 (23.76) 59 (81.94) 13 (18.06)
 Antibiotic-resistant 178 (58.75) 159 (89.33) 19 (10.67)
Location of defect 0.5323
 Proximal third 52 (17.16) 44 (84.62) 8 (15.38)
 Middle third 79 (26.07) 66 (83.54) 13 (16.46)
 Distal third 72 (23.76) 60 (83.33) 12 (16.67)
 Foot 100 (33) 90 (90.00) 10 (10.00)
Pedicle diameter <0.8 mm 85 (28.05) 72 (27.69) 13 (30.23) 0.7312
No. of perforators 0.9638
 1 228 (75.25) 196 (85.96) 32 (14.04)
 2 58 (19.14) 49 (84.48) 9 (15.52)
 ≥3 17 (5.61) 15 (88.24) 2 (11.76)
Type of arterial anastomosis 0.2935
 End-to-end 106 (34.98) 94 (88.68) 12 (11.32)
 End-to-side 197 (65.02) 166 (84.26) 31 (15.74)
No. of venous anastomoses 0.0423
 1 113 (37.29) 91 (80.53) 22 (19.47)
 ≥2 190 (62.71) 169 (88.95) 21 (11.05)
*The t test was used for continuous variables and the χ2 test was used for categorical variables.
Analysis was conducted after log transformation.

In contrast, when the median flap size was analyzed according to the number of harvesting perforators and occurrence of partial loss, the size of the partial loss group was wider than that of the non–partial loss group regardless of harvesting perforator numbers [partial loss group with one perforator, 157 cm2 (interquartile range, 157 to 319 cm2); non–partial loss group with one perforator, 132 cm2 (interquartile range, 84 to 198 cm2); partial loss group with two perforators, 165 cm2 (interquartile range, 96 to 243 cm2); non–partial loss group with two perforators, 150 cm2 (interquartile range, 90.5 to 240 cm2); partial loss group with more than three perforators, 165 cm2 (interquartile range, 352.5 to cm2); non–partial loss group with more than three perforators, 170 cm2 (interquartile range, 136 to 390 cm2)]. Harvesting the flap with one or two perforators showed a statistically significant difference between partial loss cases and non–partial loss cases (both p < 0.001). However, there was no statistical difference when more than three perforators were harvested (p = 0.098) (Table 3).

Table 3. - Comparison of Mean Flap Size by the Number of Perforators and Occurrence of Partial Loss*
Characteristic Overall (IQR) Non–Partial Loss (IQR) Partial Loss (IQR) p
No. of perforators
 1 135 (88–200) 132 (84–198) 157 (120–319) <0.001
 2 161 (96–243) 150 (90.5–240) 165 (96–243) <0.001
 ≥3 170 (148–390) 170 (136–390) 165 (352.5) 0.098
IQR, interquartile range.
*Values are in square centimeters. Analysis was performed using Mann-Whitney U test.

Finally, the mean flap size of the supradeep fat group was the largest (188.34 ± 98.39 cm2), whereas that of the suprafascial group was the smallest (155.28 ± 114.84 cm2) and there were statistical differences between the three groups (p = 0.011). After the post hoc analysis, statistical differences were observed only between the subfascial layer group and the supradeep fat layer group (p = 0.006) (Table 4).

Table 4. - Flap Size Difference for Each Flap Elevation Layer*
Characteristic Subfascial Suprafascial Supradeep Fat p
No. of reconstruction cases 180 29 94
Median flap size (IQR), cm2 135 (87–220) 120 (88–160) 175.5 (120–250) 0.011
IQR, interquartile range.
*Analysis was performed using the Kruskal-Wallis test.

Comparison of the Patient’s Laboratory Findings

Regarding the laboratory results of the patients, mean preoperative C-reactive protein level of the partial loss group was 3.08 ± 4.20 mg/dl; meanwhile, that of the non–partial loss group was 1.75 ± 3.40 mg/dl, which was statistically significant (p = 0.0073). Moreover, in the non–partial loss group, the preoperative albumin level was 3.19 ± 0.82 g/dl, which is less than that in the partial loss group (3.58 ± 0.63 g/dl), which was statistically different (p = 0.0053). No statistical significance was observed in postoperative albumin level, preoperative and postoperative glucose levels, and systolic and diastolic blood pressure levels (Table 5).

Table 5. - Preoperative and Postoperative Laboratory Findings*
Characteristic Overall No Partial Loss Partial Loss p
Preoperative
 Mean CRP ± SD, mg/dl 1.95 ± 3.56 1.75 ± 3.40 3.08 ± 4.20 0.0073
 Mean albumin ± SD, g/dl 3.52 ± 0.67 3.58 ± 0.63 3.19 ± 0.82 0.0053
 Mean glucose ± SD, mg/dl 120.54 ± 40.48 121.20 ± 41.52 116.50 ± 33.88 0.6705
Postoperative
 Mean CRP ± SD, mg/dl 2.95 ± 3.41 2.90 ± 3.40 3.21 ± 3.51 0.3566
 Mean albumin ± SD, g/dl 2.99 ± 0.47 3.02 ± 0.45 2.83 ± 0.59 0.0523
 Mean glucose ± SD, mg/dl 120.45 ± 38.33 120.30 ± 36.73 121.20 ± 47.19 0.8362
 Mean SBP for 5 days ± SD, mmHg 116.80 ± 12.40 116.40 ± 11.56 119.10 ± 16.56 0.3172
 Mean DBP for 5 days ± SD, mmHg 71.86 ± 8.31 71.64 ± 7.73 73.16 ± 11.18 0.395
CRP, C-reactive protein; SBP, systolic blood pressure; DBP, diastolic blood pressure.
*Analysis was performed using t test.

Partial Loss Area Analysis: Flap Size and Distance from the Perforator to the Necrotic Margin

The average of the distances between the perforator and the designed flap margin in the partial loss group was 12.058 ± 4.083 cm, whereas that in the non–partial loss group was 9.002 ± 3.376 cm, which was statistically significant (p < 0.0001). In addition, the mean shortest distance from the perforator to the partial loss margin was 7.767 ± 4.306 cm (Table 6).

Table 6. - Average Distances from the Perforator to the Designed Flap Margin and from the Perforator to the Partial Loss Margin*
Characteristic Overall No Partial Loss Partial Loss p
Longest distance from the perforator to flap to the margin ± SD, cm 9.439 ± 3.641 9.002 ± 3.376 12.058 ± 4.083 <0.0001
Shortest distance from the perforator to the partial loss margin ± SD, cm 7.767 ± 4.306 n/a 7.767 ± 4.306 n/a
n/a, not applicable.
*Analysis was performed using the t test.

Meanwhile, in receiver operating characteristic curve analysis, regarding the relationship between the distance from the perforator to the flap margin and partial loss, the cutoff value was 10.25 cm and the area under the curve was 0.718 (sensitivity, 0.651; specificity, 0.704; 95 percent CI, 0.634 to 0.802). In the case of the relationship between the flap size and partial loss, the cutoff value was calculated as 110 cm2 and the area under the curve was 0.636 (sensitivity, 0.884; specificity, 0.350; 95 percent CI, 0.553 to 0.719) (Fig. 3).

F3
Fig. 3.:
Receiver operating characteristic curve analysis. (Left) Correlation between partial flap loss and the longest distance from the perforator to the flap margin. (Right) Correlation between partial flap loss and flap size.

Rate of Partial Flap Necrosis According to Distance from the Perforator

When the distance between the perforator and the flap tissue was 8 cm, the incidence of partial loss was 10.3 percent (20 of 195 flaps). Moreover, when the distance was 12 cm, the incidence of partial loss was 22.9 percent (19 of 83 flaps). As the distance between the flap tissue and the perforator increased, the incidence of partial loss also increased (Table 7 and Fig. 4).

Table 7. - Incidence Rate of Partial Necrosis by the Distance from the Perforator to the Flap
Characteristic Distances from the Perforator to the Flap (cm)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Partial necrosis (a) 3 4 4 5 8 8 15 18 20 18 19 19 19 16 11 8 4 4 4 3 3 2
No. of flaps including the section (b) 303 302 302 301 296 280 261 230 195 168 136 102 83 61 43 32 17 7 5 4 4 2
Occurrence ratio (a/b) 0.010 0.013 0.013 0.017 0.027 0.029 0.057 0.078 0.103 0.107 0.140 0.186 0.229 0.262 0.256 0.250 0.235 0.571 0.800 0.750 0.750 1.000

F4
Fig. 4.:
Incidence rate of partial necrosis based on the distance from the perforator to the flap. As the distance of the flap from the perforator increases, the partial flap loss rate tends to increase.

Independent Factors of Partial Necrosis

On multivariate logistic regression analysis, the elevation layer, flap size, and preoperative C-reactive protein and albumin levels were the independent factors increasing the occurrence of partial flap losses. Based on the subfascial layer elevation, the odds ratio of the suprafascial layer is 2.476 (95 percent CI, 0.798 to 7.682) and the odds ratio of the supradeep fat layer is 3.952 (95 percent CI, 1.803 to 8.613), and that of the supradeep fat layer is statistically significant (p = 0.001). As the area of the flap increased by 10 cm2, the probability of partial flap loss occurrence increased by 4.1 percent (OR, 1.041; 95 percent CI, 1.010 to 1.072; p = 0.017).

When the preoperative C-reactive protein level increased by 1 mg/dl, the risk of partial flap loss increased by 8.4 percent (OR, 1.084; 95 percent CI, 1.004 to 1.17; p = 0.0392), and when the preoperative albumin level increased by 1 g/dl, the probability of partial flap loss occurrence was reduced to less than half (OR, 0.438; 95 percent CI, 0.271 to 0.709; p = 0.0008). The number of vein anastomoses and postoperative albumin levels, which showed statistical significance in the univariate analysis, did not show statistical significance in the multivariate analysis (Table 8).

Table 8. - Multivariate Analysis Using a Logistic Regression Model*
Characteristic OR LCI UCI p
Elevation layer
 Subfascial Ref
 Suprafascial 2.476 0.798 7.682 0.117
 Supradeep fat 3.952 1.813 8.613 0.001
Elevation layer (subfascial – suprafascial – supradeep fat) 1.547 1.196 2.002 0.001
Flap size (per 10-cm2 increase) 1.041 1.010 1.072 0.017
Preoperative CRP (per 1-mg/dl increase) 1.025 0.932 1.128 0.608
Preoperative albumin (per 1-g/dl increase) 0.603 0.344 1.059 0.078
CRP, C-reactive protein; LCI, lower 95% CI; UCI, upper 95% CI; Ref, reference.
*The backward elimination method was used among variables, which have a value of p < 0.1 on univariate analysis.

Discussion

Flap success does not mean survival of all flap regions. Unlike partial necrosis in head and neck reconstruction, which results in a fistula or leakage, immediate coverage of the defect is occasionally not necessary in lower limb reconstruction if bone and tendon are not exposed. Moreover, there was little demand to make perfect aesthetic reconstruction such as in breast reconstruction. In addition, the surviving flap region provides sufficient oxygenation and nutrition to adjacent tissues, which induces the proliferation of granulation tissue and helps overcome infection.17,18 For that reason, secondary intention and skin grafting are commonly performed to restore the partial loss area.

Although partial flap necrosis is not fatal compared to total failure, defects in the lost area still cause physical and mental burden to the patients and medical staff. Thus, from the surgeon’s perspective, knowing in advance whether there is a difference in the partial loss occurrence rate according to the distance of the flap from the perforator or the elevation plane can minimize the occurrence of partial necrosis by using an ideal flap design. This study was conducted with that expectation.

In the study, partial flap loss occurred in 14.19 percent of the 303 patients who underwent lower extremity reconstruction, resulting in three to five times more than the total loss rate of previous study, and occurred twice as much as 6 percent, which is the partial loss rate using the existing meta-analysis.11,19 To reduce partial flap loss, the following points should be considered during the elevation of the flap (Table 9).

Table 9. - Recommendation to Minimize Partial Necrosis after Anterolateral Thigh Free Flap
Characteristic Recommendation
Size of the flap Distance between the perforator and the distal margin of the flap should not exceed 12 cm
Layer of the flap Elevate the flap to the subfascial plane if bone or prosthesis coverage is required
Thinning procedure of the flap Thinning procedure needs to be avoided; if thinning is necessary, consider preserving 6–8 cm fat around the perforator, and minimize the linking vessel injury by wearing loupe magnification during trimming
No. of perforators Harvesting multiple perforators; consider inner flap anastomosis (turbocharging or supercharging)
Preoperative status of the patients Normalizing preoperative CRP and albumin level
CRP, C-reactive protein.

First, we should consider the distance from the perforator to the distal margin of the designed flap. In receiver operating characteristic curve analysis, the cutoff value was 10.55 cm, which increases the incidence of partial flap loss. In addition, when the distance between the perforator and the distal margin of the flap was 8 cm or more, partial flap loss occurred in more than 10 percent, and when it was 12 cm or more, partial flap loss occurred in more than 20 percent. Therefore, it is necessary to use the tissue that is close to the perforator in areas that must achieve reconstruction success, such as bone exposure.

When Michel Saint-Cyr mentioned the principle of flap design and skin paddle orientation, the direction of the linking vessel of the extremities was axial and that of the trunk was perpendicular to the midline.10 However, some vessels such as the superficial circumflex iliac and superficial inferior epigastric arteries, have the specific axial pattern.20 In these cases, the results of this study cannot be applied.

Second, we should consider the flap elevation layer. It was found that the flap raised to the supradeep fat plane was 5.448 times more likely to cause partial loss than the flap raised to the subfascial plane, and the comparison between the suprafascial and subfascial layers showed that the subfascial layer had a lower risk of partial flap loss. These results suggest that in the case of flap elevation to the supradeep fat layer, indirect flow through a single perforator and the subdermal plexus may not provide sufficient circulation to all areas of the flap. Alternatively, in the case of subfascial elevation, numerous microscopic capillaries and venules attached to the deep fascia in addition to the main pedicle can contribute to the circulation of the flap. Therefore, in areas where bone exposure or prosthesis coverage is required, it is necessary to elevate the flap to the subfascial plane to reduce the possibility of partial loss.

A flap thinning procedure is another method to overcome bulkiness of the flap. Although the procedure was not performed in this study, it can also induce partial flap necrosis. According to the systematic review by Sharabi et al., 25.93 percent of thinned flaps that exceeded 150 cm2 have experienced partial necrosis.21 Therefore, it is recommended to avoid flap thinning as the size of the flap increases. However, if thinning is necessary, 6 to 8 cm of fat around the perforator should be preserved, and linking vessel injury should be minimized by wearing loupe magnification during trimming, as Viviano et al. mentioned.22

The number of perforators harvested may also affect the incidence of partial necrosis. As shown in the Results section, larger flaps were elevated when more perforators were harvested; however, the incidence rate of partial necrosis among the three groups was not statistically different. In addition, the mean flap size of the partial loss group was larger than the non–partial loss group regardless of the number of the perforators. Although the mean flap size of harvesting three or more perforators did not show statistical meaning, it may be caused by the small size of the group. When harvesting multiple perforators during flap elevation, the distance from the perforator to the distal flap margin can be decreased; therefore, the incidence of partial necrosis can also be decreased.

Inner flap anastomosis (also called intraflap anastomosis, or turbocharging) is a powerful technique to reconstruct huge free flaps without necrosis.23–28 However, definition of “a large flap” is not established, and a subjective perspective is involved for determination. Through this study, we tried to investigate the circulation area that a single perforator supplies to anterolateral thigh flaps. According to the result, flap tissues more than 12 cm away from the perforator may have a necrosis rate of more than 20 percent, and it could be a criterion for defining large flaps and for whether inner flap anastomosis should be performed.

The other factor to consider is the size of the flap. Multivariate analysis showed that as the area increased by 10 cm2, more partial necrosis occurred. A huge flap requires large amounts of blood supply and the ability to deliver blood flow to distant tissue. Therefore, elevating a flap that is too wide should be avoided.

In addition to the three factors to consider when designing an anterolateral thigh free flap for lower limb reconstruction, normalizing the patient’s preoperative C-reactive protein and albumin levels before surgery will further reduce the occurrence of partial flap loss after surgery.

The effect of postoperative infection on increasing necrosis rates of pedicled and free flaps has already been reported in several studies.29–31 In this study, the degree of systemic infection or inflammation was indirectly estimated through the C-reactive protein level. As a result, the level in the partial loss group was significantly higher preoperatively than in the non–partial loss group. This may be evidence that it is important to detect and control infection and inflammation before surgery. Flap necrosis can be triggered when inflammation cell infiltration and the resulting coagulative necrosis are combined with the ischemic condition, which lasts for 4 to 7 days, which is before completion of inosculation with the recipient bed after flap insetting.

Zhou et al. reported the use of calcitriol, which has anti-inflammatory properties, and which significantly reduced partial loss of random skin flaps in rats.32 It also supports that preoperative infection and inflammation must be normalized. In addition, when free flap reconstruction was performed after head and neck cancer ablation, there was a report that low prealbumin levels significantly decreased the flap survival rate.33 Although the albumin level does not represent recent nutritional status compared to the prealbumin level because of its longer half-life, it may be a better indicator of chronic nutritional deficiency, considering that most of the subjects in this study had an American Society of Anesthesiologists score of 1 or 2. Therefore, sufficient nutritional support is needed after measuring the albumin level before surgery.

This study has the following limitations. First, it involved a relatively small population. Although the study can be considered a large-scale study, which is based on 303 patients using only anterolateral thigh free flaps for 12 years in a single institution, 43 cases of partial flap loss was relatively small; therefore, we could not perform subgroup analyses on each cause of the defects and “the difference between the mean size of each elevated flap layer and the incidence of partial necrosis.”

Another limitation was that the data analysis in this study excluded drug administration history such as prostaglandin E1 or lower-molecular-weight heparin. This is because the protocol of drug use changed during the study period. Because the effectiveness of these drugs is still unclear, it is possible to draw meaningful conclusions by collecting large-scale data from patients who have undergone surgery with a unified drug administration protocol.

CONCLUSIONS

We observed a 14.19 percent rate of partial necrosis in 303 lower limb reconstructions using anterolateral thigh free flaps. The factors increasing the risk of partial necrosis were the distance from the perforator to the distal margin of the flap, supradeep fat elevation, and flap size; in addition, preoperative C-reactive protein and albumin level also influenced the occurrence of partial loss. Taking this into consideration, if the patient is sufficiently managed before surgery and an anterolateral thigh free flap of appropriate size and thickness is raised, the possibility of partial necrosis of the flap occurring can be minimized.

PATIENT CONSENT

The patient provided written informed consent for use of her images.

REFERENCES

1. Koshima I, Soeda S. Inferior epigastric artery skin flaps without rectus abdominis muscle. Br J Plast Surg. 1989;42:645–648.
2. Taylor GI, Palmer JH. The vascular territories (angiosomes) of the body: Experimental study and clinical applications. Br J Plast Surg. 1987;40:113–141.
3. Huntley SR, Lee S, Kalra R, McGwin G, Naranje S, Shah A. Associations between season and surgical site infections in orthopaedic foot and ankle surgery. Foot (Edinb.). 2018;37:61–64.
4. Taylor GI, Corlett RJ, Dhar SC, Ashton MW. The anatomical (angiosome) and clinical territories of cutaneous perforating arteries: Development of the concept and designing safe flaps. Plast Reconstr Surg. 2011;127:1447–1459.
5. Suami H, Taylor GI, Pan WR. Angiosome territories of the nerves of the lower limbs. Plast Reconstr Surg. 2003;112:1790–1798.
6. Taylor GI, Pan WR. Angiosomes of the leg: Anatomic study and clinical implications. Plast Reconstr Surg. 1998;102:599–616; discussion 617–618.
7. Hong JP, Choi DH, Suh H, et al. A new plane of elevation: The superficial fascial plane for perforator flap elevation. J Reconstr Microsurg. 2014;30:491–496.
8. Mohan AT, Zhu L, Morsy M, et al. Reappraisal of perforasomes of the superficial femoral, descending genicular, and saphenous arteries and clinical applications to locoregional reconstruction. Plast Reconstr Surg. 2019;143:613e–627e.
9. Suh HS, Oh TS, Lee HS, et al. A new approach for reconstruction of diabetic foot wounds using the angiosome and supermicrosurgery concept. Plast Reconstr Surg. 2016;138:702e–709e.
10. Saint-Cyr M, Wong C, Schaverien M, Mojallal A, Rohrich RJ. The perforasome theory: Vascular anatomy and clinical implications. Plast Reconstr Surg. 2009;124:1529–1544.
11. Xiong L, Gazyakan E, Kremer T, et al. Free flaps for reconstruction of soft tissue defects in lower extremity: A meta-analysis on microsurgical outcome and safety. Microsurgery. 2016;36:511–524.
12. Gabrysz-Forget F, Tabet P, Rahal A, Bissada E, Christopoulos A, Ayad T. Free versus pedicled flaps for reconstruction of head and neck cancer defects: A systematic review. J Otolaryngol Head Neck Surg. 2019;48:13.
13. Song YG, Chen GZ, Song YL. The free thigh flap: A new free flap concept based on the septocutaneous artery. Br J Plast Surg. 1984;37:149–159.
14. Rozen WM, Grinsell D, Koshima I, Ashton MW. Dominance between angiosome and perforator territories: A new anatomical model for the design of perforator flaps. J Reconstr Microsurg. 2010;26:539–545.
15. Wu CC, Lin PY, Chew KY, Kuo YR. Free tissue transfers in head and neck reconstruction: Complications, outcomes and strategies for management of flap failure. Analysis of 2019 flaps in single institute. Microsurgery. 2014;34:339–344.
16. Lee ZH, Abdou SA, Ramly EP, et al. Larger free flap size is associated with increased complications in lower extremity trauma reconstruction. Microsurgery. 2020;40:473–478.
17. Patzakis MJ, Zalavras CG. Chronic posttraumatic osteomyelitis and infected nonunion of the tibia: Current management concepts. J Am Acad Orthop Surg. 2005;13:417–427.
18. Rimmer RA, Duffy AN, Knops AM, et al. The role of free tissue transfer in the management of chronic frontal sinus osteomyelitis. Laryngoscope. 2019;129:1497–1504.
19. Chang EI, Zhang H, Liu J, Yu P, Skoracki RJ, Hanasono MM. Analysis of risk factors for flap loss and salvage in free flap head and neck reconstruction. Head Neck. 2016;38(Suppl 1):E771–E775.
20. Suh HSP, Jeong HH, Choi DH, Hong JPJP. Study of the medial superficial perforator of the superficial circumflex iliac artery perforator flap using computed tomographic angiography and surgical anatomy in 142 patients. Plast Reconstr Surg. 2017;139:738–748.
21. Sharabi SE, Hatef DA, Koshy JC, Jain A, Cole PD, Hollier LH Jr. Is primary thinning of the anterolateral thigh flap recommended? Ann Plast Surg. 2010;65:555–559.
22. Viviano SL, Liu FC, Therattil PJ, Lee ES, Keith JD. Peripheral pruning: A safe approach to thinning extra-large anterolateral thigh flaps. Ann Plast Surg. 2018;80(Suppl 4):S164–S167.
23. Deleyiannis FW, Badeau AM, Leem TH, Song JI. Supercharging and augmenting venous drainage of an anterolateral thigh free flap: Options and indications. Plast Reconstr Surg Glob Open. 2014;2:e135.
24. Hung SY, Loh CYY, Chen HC. Supercharging extra-large anterolateral thigh flaps for single-stage resurfacing of massive burn defects over upper extremity elbow joints. J Burn Care Res. 2018;39:831–834.
25. Lee YJ, Lee YJ, Oh DY, Jun YJ, Rhie JW, Moon SH. Reconstruction of wide soft tissue defects with extended anterolateral thigh perforator flap turbocharged technique with anteromedial thigh perforator. Microsurgery. 2020;40:440–446.
26. Vijayan R, Al-Aswad F, Kannan RY. Supermicrosurgery-assisted venous supercharging of a reverse-flow angular artery perforator flap for nasal reconstruction. J Plast Reconstr Aesthet Surg. 2017;70:281–283.
27. Wang X, Pan J, Xiao D, et al. Comparison of arterial supercharging and venous superdrainage on improvement of survival of the extended perforator flap in rats. Microsurgery. 2020;40:874–880.
28. Lee G, Pourmoussa AJ, Perrault D, Wong AK. Supercharged free transverse rectus abdominis myocutaneous flap: An autologous reconstructive option for the thin breast reconstruction patient. Cureus. 2020;12:e8776.
29. Gong X, Cui J, Jiang Z, Lu L, Li X. Risk factors for pedicled flap necrosis in hand soft tissue reconstruction: A multivariate logistic regression analysis. ANZ J Surg. 2018;88:E127–E131.
30. Yan H, Yang M, Lineaweaver WC, Myers RS, Chen H, Zhang F. Flap reconstruction of distal lower extremity wounds in diabetic patients. Ann Plast Surg. 2011;67:426–428.
31. Wax MK, Rosenthal E. Etiology of late free flap failures occurring after hospital discharge. Laryngoscope. 2007;117:1961–1963.
32. Zhou KL, Zhang YH, Lin DS, Tao XY, Xu HZ. Effects of calcitriol on random skin flap survival in rats. Sci Rep. 2016;6:18945.
33. Shum J, Markiewicz MR, Park E, et al. Low prealbumin level is a risk factor for microvascular free flap failure. J Oral Maxillofac Surg. 2014;72:169–177.
Copyright © 2022 by the American Society of Plastic Surgeons