Figures
Abstract
Objective
The purpose of this study is to determine the correlation between non-technical risk factors and the perioperative flap survival rate and to evaluate the choice of skin flap for the reconstruction of foot and ankle.
Methods
This was a clinical retrospective study. Nine variables were identified. The Kaplan-Meier method coupled with a log-rank test and a Cox regression model was used to predict the risk factors that influence the perioperative flap survival rate. The relationship between postoperative wound infection and risk factors was also analyzed using a logistic regression model.
Results
The overall flap survival rate was 85.42%. The necrosis rates of free flaps and pedicled flaps were 5.26% and 20.69%, respectively. According to the Cox regression model, flap type (hazard ratio [HR] = 2.592; 95% confidence interval [CI] (1.606, 4.184); P < 0.001) and postoperative wound infection (HR = 0.266; 95% CI (0.134, 0.529); P < 0.001) were found to be statistically significant risk factors associated with flap necrosis. Based on the logistic regression model, preoperative wound bed inflammation (odds ratio [OR] = 11.371,95% CI (3.117, 41.478), P < 0.001) was a statistically significant risk factor for postoperative wound infection.
Conclusion
Flap type and postoperative wound infection were both independent risk factors influencing the flap survival rate in the foot and ankle. However, postoperative wound infection was a risk factor for the pedicled flap but not for the free flap. Microvascular anastomosis is a major cause of free flap necrosis. To reconstruct complex or wide soft tissue defects of the foot or ankle, free flaps are safer and more reliable than pedicled flaps and should thus be the primary choice.
Citation: Li X, Cui J, Maharjan S, Lu L, Gong X (2016) Reconstruction of the Foot and Ankle Using Pedicled or Free Flaps: Perioperative Flap Survival Analysis. PLoS ONE 11(12): e0167827. https://doi.org/10.1371/journal.pone.0167827
Editor: Fabio Santanelli, di Pompeo d'Illasi, di Pompeo d'Illasi, Universita degli Studi di Roma La Sapienza Facolta di Medicina e Psicologia, ITALY
Received: March 18, 2016; Accepted: November 21, 2016; Published: December 8, 2016
Copyright: © 2016 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: We have not received any funding or sources of support.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The foot and ankle are prone to injuries and diseases because of insufficient soft tissue [1]. Complex soft tissue defects of the foot and ankle caused by trauma, infection, tumor cancer or diabetes are common and can be accompanied by exposed tendons, neurovascular bundles and bone. There are multiple options for the reconstruction of complex soft tissue defects in these areas, including the use of both pedicled flaps and free flaps (e.g., a lateral supramalleolar flap [2], a peroneal artery perforator flap [3,4], posterior tibial artery perforator flap [5], a sural neurocutaneous/neurofasciocutaneous flap [6,7,8], a medial pedis flap [9], a dorsal metatarsal flap [10], a dorsalis pedis flap [11], a pedicled or free medial plantar flap [12,13], a free groin flap [14,15], a free anterolateral thigh perforator flap [16,17,18], or a free anteromedial thigh perforator flap [19]). In addition, the successful reconstruction of complex soft tissue defects of the foot and ankle is critical because of the unique standing, weight-bearing and walking functions of the foot. The management and reconstruction of soft tissue defects must thus be a surgeon’s primary focus.
Although flap transfers have become the primary method of reconstruction of complex soft tissue defects of the foot and ankle and although microsurgical techniques have advanced, skin flap failure often occurs during the perioperative period. Once skin flap failure has occurred, it can have devastating consequences. When selecting a skin flap, several factors must be considered, such as the pliability of the skin flap, the stability of standing and walking, donor site morbidity, the vascular condition at the recipient site [20], the size of the soft tissue defect, and the flap survival rate.
The purpose of this study is to determine the correlation between non-technical risk factors and the perioperative flap survival rate and to evaluate the choice of skin flap for the reconstruction of foot and ankle. In this study, the perioperative period was defined as within 2 weeks after flap transfer.
Patients and Methods
Data Collection Criteria and Study Design
This retrospective study was approved by the institutional review committee and ethics committee at the First Hospital of Jilin University. Written informed consent to participate in this study was obtained from each patient. The inclusion criteria were patients who had a unilateral foot or ankle injury with complex soft tissue defects and who had undergone various pedicled or free flap procedures. Only the initial pedicled or free flap procedure and its complications were studied. Patients were excluded if they had a bilateral foot or ankle injury because repeated data analysis can increase the sample error. Although secondary flap procedures were performed, the secondary transplanted flap and its complications were not the focus of our study. Considering the above-mentioned criteria, we carefully reviewed hospital records and found that 144 patients met these criteria between February 2007 and December 2014.
The characteristics of these patients, the flap transfer procedures and postoperative complications were recorded and researched. The nine risk factors assessed in this study were as follows: patient age (≤ 40 years, > 40 years), cigarette smoking (1 = smoker, 0 = nonsmoker), hypertension (1 = Yes, 0 = No), osteomyelitis (1 = Yes, 0 = No), preoperative wound bed inflammation (1 = Yes, 0 = No), trauma activation (1 = Yes, 0 = No), anatomical region (1 = hindfoot and ankle region, 2 = midfoot region, 3 = forefoot region, 4 = multiple regions), the type of flap (1 = pedicled flap, 2 = free flap), and postoperative wound infection (1 = Yes, 0 = No). The flap outcome was subdivided into flap survival and flap necrosis (1 = flap necrosis, 0 = flap survival), which were categorical variables. If free flap failure occurred within the first 48 hours, these flaps would be excluded because their failure could be more closely related to a microvascular thrombosis. A univariate analysis of the risk factors for flap necrosis was first performed. The variables that were identified as statistically significant or nearly significant were entered into a multivariate regression analysis. Diabetes mellitus could not be analyzed as a predictor because the number of patients with diabetes mellitus in this study was insufficient.
Patient Descriptions and Management
Fourteen different flaps were used in 144 patients. Thirty patients were female, and 114 were male, with a 3.8:1 male-to-female ratio. There were 64 left-foot and 80 right-foot injuries. Comorbidities included hypertension, diabetes mellitus and osteomyelitis (Table 1). Trauma was the most common etiology, followed by skin ulcers and inflammation and then tumors (Table 1). Flap necrosis > 60% was regarded as complete necrosis (CN), and flap necrosis ≤ 60% was regarded as partial necrosis (PN) [21]. According to Godina’s study [22], the time window from trauma to 72 hours after the trauma was considered as the acute period, while the time window of the subacute stage was between 72 hours and 90 days after the trauma. In our series, 107 trauma patients underwent the flap transfer procedures in the subacute period (Table 2). In addition, preoperative wound bed inflammation and postoperative wound infection were identified by bacterial cultivation. Based on our clinical experience, the foot and ankle were divided into eight subunits within three regions (Fig 1).
K-wires, hollow screws and external fixators were used to repair fractures of the foot and ankle. Additionally, the injured tendons and major nerves were reconstructed, and negative pressure wound therapy was used. Thorough and complete debridement was carried out before the pedicled or free flap transfer. Postoperative flap care and monitoring were carried out 3–5 days after surgery, and low-molecular-weight heparin calcium was systemically administered for anticoagulation.
Statistical Analysis
The descriptive statistics were derived from the case numbers and/or the percentages of patients. Initially, the Kaplan-Meier method coupled with a log-rank test was used for the univariate analysis to predict the relationship between the perioperative flap survival rate and the risk factors. After the univariate analysis, the variables that were identified as statistically significant or nearly significant were entered into a multivariate Cox regression model to forecast the main risk factor for perioperative flap necrosis. The relationship between postoperative wound infection and the risk factors was also analyzed using a binary logistic regression model. P values < 0.05 indicated statistical significance. All statistical analyses were performed using IBM SPSS 21.0 software (IBM Corp., Armonk, N.Y., U.S.A.).
Results
Perioperative Complications
The average age at the time of flap surgery was 37.9 years (range, 3–74 years). In the trauma patient group, the mean interval between injury and flap transfer was 13.7 days (range, 8 hours to 63 days), and the rates of postoperative wound infection and flap necrosis in the subacute stage were 18.69% and 14.02%, respectively (Table 2). The flap dimensions ranged from 4×3 cm to 35×20 cm. Postoperative wound infections were observed in 27 patients. The donor sites for 94 patients underwent full-thickness skin grafts, which survived completely. Wound dehiscence at the donor site was observed in three patients. Five of 57 free flaps required re-exploration within the first 24–48 hours because of arterial thrombosis at the microvascular anastomosis site. Three free flaps were successfully salvaged, while two free flaps with arterial thrombosis experienced complete necrosis within the first 24–48 hours. In addition, one free flap exhibited PN on the sixth postoperative day (Fig 2E and 2F). Eleven pedicled flaps experienced PN, and seven pedicled flaps exhibited CN. The overall rate of pedicled and free flap survival was 85.42% (123 of 144 cases). The necrosis rates of the free flaps and pedicled flaps were 5.26% (3 of 57 cases) and 20.69% (18 of 87 cases), respectively (Table 3). Of the 21 patients with flap necrosis, a secondary flap transfer was performed in four cases, while 10 patients underwent a full-thickness skin graft, and the remaining patients healed with local wound care. In all other patients, all wounds healed without complications. Table 3 shows the results for the 144 skin flaps.
a, Mutilating forefoot injury; b, after forefoot debridement; c and d, design and harvesting of free medial plantar flap; e and f, PN of the flap on the sixth postoperative day.
Analysis of Risk Factors Associated with Flap Necrosis
Two free flap failures were excluded because their failure occurred within the first 48 postoperative hours. One hundred forty-two flaps were used to analyze the non-technical risk factors associated with flap necrosis.
Fig 3 shows the relationship between the perioperative flap survival rate and the risk factors. In the univariate analysis, preoperative wound bed inflammation (P = 0.005), anatomical region (P = 0.010), flap type (P < 0.001) and postoperative wound infection (P < 0.001) were statistically significant risk factors. A multivariate Cox regression model was used to analyze the statistically significant variables found by the univariate analysis. Based on the results of the multivariate Cox model (Table 4), flap type (hazard ratio [HR] = 2.592; 95% confidence interval [CI] (1.606, 4.184); P < 0.001) and postoperative wound infection (HR = 0.266; 95% CI (0.134, 0.529); P < 0.001) were found to be statistically significant factors. However, as shown in Table 5, in the free flap group, postoperative wound infection (HR = 0.509; 95% CI (0.184, 1.405); P = 0.192) was not a statistically significant risk factor for free flap necrosis, whereas in the pedicled flap group (Table 6), postoperative wound infection (HR = 0.202; 95% CI (0.071, 0.578); P = 0.003) was a statistically significant risk factor.
The outcomes of the univariate analysis of risk factors influencing the perioperative flap survival rate. The univariate analysis was performed using the Kaplan-Meier method coupled with a log-rank test.
Risk Factors Associated with Postoperative Wound Infection
As shown in Table 7, preoperative wound bed inflammation (odds ratio [OR] = 11.371; 95% CI (3.117, 41.478); P < 0.001) was found to be a statistically significant risk factor for postoperative wound infection. No other analyzed risk factors were found to be statistically significant.
Discussion
It was assumed that predictive variables would affect perioperative flap survival. Fig 3 indicates that preoperative wound bed inflammation, anatomical region, flap type and postoperative wound infection specifically affect the perioperative flap survival rate. In the multivariate Cox regression model (Table 4), flap type and postoperative wound infection were both independent risk factors influencing flap survival. Flap type and postoperative wound infection were associated with HRs of 2.592 and 0.266, respectively. Considering their HRs, the major risk factor influencing the perioperative flap survival rate was flap type.
Table 5 shows that postoperative wound infection is not a risk factor of free flap necrosis, whereas, Wong et al. [23] reported that operative time is a significant risk factor for free flap loss. However, in our series, five free flaps occurred arterial thrombosis within the first 24–48 hours. Among them, two free flaps exhibited CN within the first 24–48 hours. These flap failures could be more closely related to arterial thrombosis than to infection because infection can result in arterial thrombosis at a later stage (e.g., 3 days later) after complete and thorough debridement. Therefore, we agree that microvascular anastomosis is a major cause of free flap necrosis.
In contrast, as shown in Table 6, postoperative wound infection is a risk factor associated with pedicled flap necrosis. Infection gives rise to an inflammatory reaction, and the release of inflammatory mediators can lead to the formation of thrombi or vasospasms and can thus damage blood vessels [24,25]. Surgical site infection can also occur due to the complex effects of both bacterial colonization and host defense mechanisms, resulting in an increased risk of vascular surgery failure [24]. Postoperative wound infection is thus a risk factor of pedicled flap necrosis. However, Bekara et al. agree that the significant risk factors for pedicled flap complications in the lower limb include age older than 60 years, diabetes, and arteriopathy [26]. Wei et al. believe that flap-related factors may be closely related to pedicled flap necrosis; namely, the length of the pedicled flap and the width of the pedicled flap around the perforator pedicle may lead to a venous drainage problem and an insufficient arterial blood supply in a large pedicled flap by affecting the number of choke vessels between angiosomes or inclusive linking vessels between perforasomes [27]. Thus, pedicled flap necrosis is not only related to postoperative wound infection but also associated with flap-related factors.
Regarding risk factors associated with postoperative wound infection, according to Table 7, preoperative wound bed inflammation was a risk factor for postoperative wound infection. According to Godina’s study [22], the treatment of complex lower limb traumas in the acute period (< 72 hours) or in the chronic period after multiple debridements (> 90 days) reduces the risk of recipient site infection and osteomyelitis. In addition, that study reported that the postoperative infection rates for flap transfer in the acute, subacute, and chronic stages were 1.5%, 17.5% and 6%, respectively. However, in the trauma patient group in the current study, the postoperative wound infection rate in the subacute stage was 18.69% (20 of 107 cases) (Table 2), strongly suggesting the incidence of preoperative recipient site inflammation (29.17%), which was possibly responsible for postoperative wound infection (18.75%). Therefore, to increase the perioperative flap survival rate, multiple thorough debridements should be performed before flap transfer.
With regard to flap selection for reconstruction of the foot and ankle, Bekara et al. agree that the success of free flaps and pedicled flaps in reality appears similar [28]. Gir et al. believe that pedicled perforator flap use is a safe and reliable procedure for reconstructing soft tissue defects [29]. Xiong et al. [30] and Kang et al. [31] think that free flap use in foot and ankle reconstruction is safe and reliable due to very robust vascularization and a low risk of infection. However, in our study, Fig 3 and Table 4 indicate that flap type significantly affects the perioperative flap survival rate in the foot and ankle. Zhu et al. report that the overall rate of pedicled and free flap survival is 77.9% and that the necrosis rates of pedicled and free flaps are 17.7% and 7.9%, respectively [32]. Additionally, Bekara et al. indicate that the overall pedicled flap survival rate is 84.3% [26], and Hollenbeck et al. found an overall primary free flap survival rate of 92% [1]. In our series, the complete survival rate of free flaps (94.74%, 54 of 57 cases) was greater than that of pedicled flaps (79.31%, 69 of 87 cases) (Table 3). Therefore, for the reconstruction of complex or wide soft tissue defects of the foot or ankle, we agree that the use of free flaps is safer and more reliable than the use of pedicled flaps.
For soft tissue defects in the hindfoot and ankle regions, pedicled peroneal artery perforator flaps, posterior tibial artery perforator flaps and sural neurocutaneous flaps are frequently used because these flaps’ dimensions are large enough to reconstruct a single subunit. When reconstructing these subunits, either free flaps or pedicled flaps can be used. However, a study by Pinsolle et al. indicates that pedicled flaps should be the first choice when reconstructing soft tissue defects in these areas but that free flaps should be used for oversized or composite defects [33]. In the midfoot region (Fig 2), soft tissue defects of subunit VI are not common, and it is difficult to cover soft tissue defects in this subunit with a pedicled flap from the leg and dorsal pedis because of the transfer distance. Moreover, in this region, when bone is not involved, reconstruction is accomplished with a full-thickness skin graft, rather than a complex flap, due to the low functional demand. In subunit V, an isolated soft tissue defect can be reconstructed using either a free flap or a pedicled flap because of the small size of this subunit. With regard to reconstruction of the forefoot, the reconstruction must consider the unique characteristics of this region. In particular, the forefoot is situated at the most distal region of the foot, so the transfer distance for a pedicled flap from the leg and foot is too great to fully cover large defects in the forefoot region. Because this area is usually covered by a shoe, reconstruction also requires a thin flap. Therefore, we agree that a free anterolateral thigh flap is a good choice for the reconstruction of complex or wide soft tissue defects of the forefoot.
There are some limitations to this research. First, the patient population exhibited a certain heterogeneity. Second, non-technical variables related to flap loss were researched, and flap loss is associated with both flap factors and microvascular problem. Third, identifying preoperative wound bed inflammation and postoperative wound infection via bacterial cultivation is likely to produce some false positives or false negatives result. Finally, the outcome of this research requires further validation in a larger set of patients, preferably in a multi-center study.
Conclusions
Flap type and postoperative wound infection were both independent risk factors influencing the flap survival rate in the foot and ankle. However, postoperative wound infection was a risk factor for the pedicled flap but not for the free flap. In addition, pedicled flap necrosis may be associated with flap-related factors. Microvascular anastomosis is a major cause of free flap necrosis. Preoperative wound bed inflammation was found to be a risk factor associated with postoperative wound infection. To increase the perioperative flap survival rate, multiple thorough debridements should be performed before flap transfer. For the reconstruction of complex or wide soft tissue defects of the foot or ankle, free flaps are safer and more reliable than pedicled flaps and should thus be the primary choice.
Supporting Information
S1 Fig. The eight subunits (I-VIII) within three regions (forefoot, midfoot, and hindfoot and ankle) of the foot and ankle.
https://doi.org/10.1371/journal.pone.0167827.s001
(XLSX)
S2 Fig. A 49-year-old male patient suffered from a mutilating forefoot injury, and the forefoot was reconstructed using the free contralateral plantar medial flap.
a, Mutilating forefoot injury; b, after forefoot debridement; c and d, design and harvesting of free medial plantar flap; e and f, PN of the flap on the sixth postoperative day.
https://doi.org/10.1371/journal.pone.0167827.s002
(XLSX)
S3 Fig. Overall survival.
The outcomes of the univariate analysis of risk factors influencing the perioperative flap survival rate. The univariate analysis was performed using the Kaplan-Meier method coupled with a log-rank test.
https://doi.org/10.1371/journal.pone.0167827.s003
(XLSX)
S2 Table. The outcome of reconstruction of the foot and ankle in trauma patients in different periods
https://doi.org/10.1371/journal.pone.0167827.s005
(XLSX)
S4 Table. Multivariate Cox model results for predictive factors influencing flap survival rate (n = 142)
https://doi.org/10.1371/journal.pone.0167827.s007
(XLSX)
S5 Table. Multivariate Cox model results for predictive factors of free flap (n = 55) necrosis
https://doi.org/10.1371/journal.pone.0167827.s008
(XLSX)
S6 Table. Multivariate Cox model results for predictive factors of pedicled flap (n = 87) necrosis
https://doi.org/10.1371/journal.pone.0167827.s009
(XLSX)
S7 Table. Binary logistic regression model for risk factors associated with postoperative wound infection (n = 144)
https://doi.org/10.1371/journal.pone.0167827.s010
(XLSX)
Author Contributions
- Conceptualization: XL XG JC.
- Data curation: SM LL.
- Formal analysis: XL XG JC.
- Funding acquisition: None.
- Investigation: XL SM LL.
- Methodology: XL XG.
- Project administration: XL XG.
- Resources: XL XG JC.
- Software: XL SM LL.
- Supervision: SM LL.
- Validation: JC LL.
- Visualization: XL JC SM.
- Writing – original draft: XL XG.
- Writing – review & editing: XL XG JC.
References
- 1. Hollenbeck ST, Woo S, Komatsu I, Erdmann D, Zenn MR, Levin LS. Longitudinal outcomes and application of the subunit principle to 165 foot and ankle free tissue transfers. Plast Reconstr Surg. 2010; 125: 924–934. pmid:20009789
- 2. Voche P, Merle M, Stussi JD. The lateral supramalleolar flap: experience with 41 flaps. Ann Plast Surg. 2005; 54: 49–54. pmid:15613883
- 3. Lu TC, Lin CH, Lin CH, Lin YT, Chen RF, Wei FC. Versatility of the pedicled peroneal artery perforator flaps for soft-tissue coverage of the lower leg and foot defects. J Plast Reconstr Aesthet Surg. 2011; 64: 386–393. pmid:20538534
- 4. Ruan HJ, Cai PH, Schleich AR, Fan CY, Chai YM. The extended peroneal artery perforator flap for lower extremity reconstruction. Ann Plast Surg. 2010; 64: 451–457. pmid:20224342
- 5. Schaverien MV, Hamilton SA, Fairburn N, Rao P, Quaba AA. Lower limb reconstruction using the islanded posterior tibial artery perforator flap. Plast Reconstr Surg. 2010; 125: 1735–1743. pmid:20517099
- 6. Li Y, Xu J, Zhang XZ. Lowering the pivot point of sural neurofasciocutaneous flaps to reconstruct deep electrical burn wounds in the distal foot. Burns. 2013; 39: 808–813. pmid:22995425
- 7. Liu L, Zou L, Li Z, Zhang Q, Cao X, Cai J. The extended distally based sural neurocutaneous flap for foot and ankle reconstruction: a retrospective review of 10 years of experience. Ann Plast Surg. 2014; 72: 689–694. pmid:23241805
- 8. Chai Y, Zeng B, Zhang F, Kang Q, Yang Q. Experience with the distally based sural neurofasciocutaneous flap supplied by the terminal perforator of peroneal vessels for ankle and foot reconstruction. Ann Plast Surg. 2007; 59: 526–531. pmid:17992146
- 9. Yoon ES, Kim DW, Chun D, Dhong ES, Koo SH, Park SH, et al. An anatomic study and clinical application of medial pedis flap in Asians. Ann Plast Surg. 2007; 58: 517–522. pmid:17452836
- 10. Balakrishnan C, Chang YJ, Balakrishnan A, Careaga D. Reversed dorsal metatarsal artery flap for reconstruction of a soft tissue defect of the big toe. Can J Plast Surg. 2009; 17: e11–12. pmid:20808742
- 11. Zhao P, Li S. Comparison of venous drainage in flow-through and conventional dorsalis pedis flaps for repair of dorsal foot defects. Int J Clin Exp Med. 2015; 8: 9519–9524. pmid:26309617
- 12. Acikel C, Celikoz B, Yuksel F, Ergun O. Various applications of the medial plantar flap to cover the defects of the plantar foot, posterior heel, and ankle. Ann Plast Surg. 2003; 50: 498–503. pmid:12792539
- 13. Zelken JA, Lin CH. An Algorithm for Forefoot Reconstruction With the Innervated Free Medial Plantar Flap. Ann Plast Surg. 2016; 76: 221–226. pmid:26101991
- 14. Hong JP, Sun SH, Ben-Nakhi M. Modified superficial circumflex iliac artery perforator flap and supermicrosurgery technique for lower extremity reconstruction: a new approach for moderate-sized defects. Ann Plast Surg. 2013; 71: 380–383. pmid:23187712
- 15. Hsu WM, Chao WN, Yang C, Fang CL, Huang KF, Lin YS, et al. Evolution of the free groin flap: the superficial circumflex iliac artery perforator flap. Plast Reconstr Surg. 2007; 119: 1491–1498. pmid:17415243
- 16. Acar MA, Güleç A, Aydin BK, Erkoçak ÖF, Yilmaz G, Şenaran H. Reconstruction of foot and ankle defects with a free anterolateral thigh flap in pediatric patients. J Reconstr Microsurg. 2015; 31: 225–232. pmid:25629205
- 17. Sekido M, Yamamoto Y, Furukawa H, Sugihara T. Change of weight-bearing pattern before and after plantar reconstruction with free anterolateral thigh flap. Microsurgery. 2004; 24: 289–292. pmid:15274185
- 18. Liu L, Cao X, Zou L, Li Z, Cao X, Cai J. Extended anterolateral thigh flaps for reconstruction of extensive defects of the foot and ankle. PLoS One. 2013; 8: e83696. pmid:24376731
- 19. Schoeller T, Huemer GM, Shafighi M, Gurunluoglu R, Wechselberger G, Piza-Katzer H. Free anteromedial thigh flap: clinical application and review of literature. Microsurgery. 2004; 24: 43–48. pmid:14748024
- 20. Battiston B, Antonini A, Tos P, Daghino W, Massazza G, Riccio M. Microvascular reconstructions of traumatic-combined tissue loss at foot and ankle level. Microsurgery. 2011; 31: 212–217. pmid:21351140
- 21. Gupta A, Lakhiani C, Lim BH, Aho JM, Goodwin A, Tregaskiss A, et al. Free tissue transfer to the traumatized upper extremity: Risk factors for postoperative complications in 282 cases. J Plast Reconstr Aesthet Surg. 2015; 68: 1184–1190. pmid:26212638
- 22. Godina M. Early microsurgical reconstruction of complex trauma of the extremities. Plast Reconstr Surg. 1986; 78: 285–292. pmid:3737751
- 23. Wong AK, Joanna NT, Peric M, Shahabi A, Vidar EN, Hwang BH, et al. Analysis of risk factors associated with microvascular free flap failure using a multi-institutional database. Microsurgery. 2015; 35: 6–12. pmid:24431159
- 24. Inui T, Bandyk DF. Vascular surgical site infection: risk factors and preventive measures. Semin Vasc Surg. 2015; 28: 201–207. pmid:27113288
- 25. Lin YT, Su ST, Lo S, Hu CH, Lin CH, Wei FC. Risk factors for reexploration in toe-to-hand transfer: a multivariate analysis of 363 cases. Plast Reconstr Surg. 2015; 135: 501–506. pmid:25357159
- 26. Bekara F, Herlin C, Mojallal A, Sinna R, Ayestaray B, Letois F, et al. A Systematic Review and Meta-Analysis of Perforator-Pedicled Propeller Flaps in Lower Extremity Defects: Identification of Risk Factors for Complications. Plast Reconstr Surg. 2016; 137: 314–331. pmid:26371391
- 27. Wei JW, Ni JD, Dong ZG, Liu LH, Liu X, Peng P. Letter: A Systematic Review and Meta-Analysis of Perforator-Pedicled Propeller Flaps in Lower Extremity Defects: Identification of Risk Factors for Complications. Plast Reconstr Surg. 2016.
- 28. Bekara F, Herlin C, Somda S, de Runz A, Grolleau JL, Chaput B. Free versus perforator-pedicled propeller flaps in lower extremity reconstruction: What is the safest coverage? A meta-analysis. Microsurgery. 2016.
- 29. Gir P, Cheng A, Oni G, Mojallal A, Saint-Cyr M. Pedicled-perforator (propeller) flaps in lower extremity defects: a systematic review. J Reconstr Microsurg. 2012; 28: 595–601. pmid:22715046
- 30. Xiong L, Gazyakan E, Kremer T, Hernekamp FJ, Harhaus L, Saint-Cyr M, et al. Free flaps for reconstruction of soft tissue defects in lower extremity: A meta-analysis on microsurgical outcome and safety. Microsurgery. 2016.
- 31. Kang MJ, Chung CH, Chang YJ, Kim KH. Reconstruction of the lower extremity using free flaps. Arch Plast Surg. 2013; 40: 575–583. pmid:24086813
- 32. Zhu YL, Wang Y, He XQ, Zhu M, Li FB, Xu YQ. Foot and ankle reconstruction: an experience on the use of 14 different flaps in 226 cases. Microsurgery. 2013; 33: 600–604. pmid:24038123
- 33. Pinsolle V, Reau AF, Pelissier P, Martin D, Baudet J. Soft-tissue reconstruction of the distal lower leg and foot: are free flaps the only choice? Review of 215 cases. J Plast Reconstr Aesthet Surg. 2006; 59: 912–917; discussion 918. pmid:16920580