Recently, numerous studies have showed that extensive DM foot ulcer could be safely treated with free flap transfer [8–10]. A systematic review study based on 18 studies of free tissue transfer for ischemic lower limb wound coverage showed a high rate of free flap survival as 92% and limb salvage rate of 83.4% [11]. In 1985 Briggs et al. firstly described the concept of combining vascular bypass surgery and free tissue transfer which could effectively manage the ischemic limbs as well as covered the complex defects [4]. This method expressively increased the salvage rate of the critical ischemic diabetic limb. Subsequently, several studies on combination between vascular distal bypass surgery and free flap transfer have been performed [12–15]. Although bypass surgery was successfully used for chronic limb ischemia management earlier on, PTA has slowly superseded open surgical bypass because of its advantages included reducing complication and length of hospitalization, feasible in elder and patients with poor general condition or even in patients with severe multi-segmental occlusive disease [16–18]. In addition, with the introduction of small-scale balloon and drug-coated balloon, PTA is significantly effective with better primary patency and low rate of restenosis [19]. Our institute has recently employed PTA as the first-line treatment for infrainguinal PAD patients. In the present study, we combined PTA and free flap surgery for the ischemic chronic wounds in diabetic patients and the overall flap survival rate was 89.1% comparable to the previous studies. Although the flap survival rate and limb salvage rate were relatively high, Oh et al. reported some obstacle factors included PAD, history of previous angioplasty and using immunosuppressive agents after renal transplant [20]. Authors emphasized that PAD and having a history of angioplasty had a significantly higher risk of flap failure (17.59 times and 10.212 times respectively). Twenty-three patients in our study were grade 4 and 5 PAD patients with severe stenosis in at least one main of infrapopliteal arteries. Though four of five flaps loss were among severe PAD patients, the analysis showed no significant difference between the severe PAD patients and the other patients. All of the severe PAD patients in this study underwent PTA before the flap transfer surgery to achieve the straight-line blood flow to the pedal arch, however, we failed to reconstruct the pedal arch in one patient classified as type 3. Despite having no impact on to amputation rate, the quality of pedal arch had affected the rate of wound healing and time to healing as well [7, 21]. In this series, we also found the crucial influence of quality of pedal arch to the free flap survival. Among 13 patients with complete final pedal arch after PTA procedure, there was no flap necrosis, while four flaps loss were found in the group of ten patients with compromised pedal arch.
Hong stated that the evaluation of recipient arteries perfusion and identification of an appropriate vessel may be the major challenge in reconstructive microsurgery [3]. Previous literatures showed that the appropriate recipient artery could be selected preoperatively via angiography or CT angiography images [6, 22, 23]. In addition, Kim et al. employed Doppler ultrasonography to check the recipient artery and demonstrated that over 40 cm/s of peak blood flow velocity (PBFV) should be achieved to assure the success in free tissue transfer [24]. Nevertheless, it was indicated that pre-operation blood flow measurement may not be similar to actual findings in surgery [3]. In our institute, we did not measure PBFV before the surgery but we preferred using CT angiography as a minimally invasive modality for vascular status assessment and flap planning as well. However, the final decision must be made based on the intraoperative condition of the arteries. The surgery would be abandoned if no sufficient vessel were found.
Based on the evidence of the safe utility of revascularized recipient artery, we also anastomosed the flaps to 18 arteries that had undergone PTA [22, 25]. Although three of 18 flaps totally necrosed, no significant correlation was found between the revascularized recipient artery and the flap loss rate.
End-to-side anastomosis was performed in 40 flaps because it was believed that preservation of major arteries had been very important in CLI patients. In addition, the flaps were also occasionally supplied by the retrograde blood flow, especially in case of intact pedal arch. Lee et al. found a significantly lower risk of wound complication of end-to-side anastomosis rather than end-to-end when they performed a study of factors affect the flap survival and complication [9]. Though heavy calcification of the arterial walls was common in PAD patients, we always tried to find the least atherosclerotic for anastomosis. The end-to-end modality would only be used if we could not find less diseased area or unsatisfied result after end-to-side anastomosis.
There was no consensus in using fasciocutaneous or muscle flap combined with skin graft. We believed that fasciocutaneous was sufficient for the diabetic foot ulcers often thin and superficial. In this study, ALT was our most common flap and MSAP flap was used as an alternative especially for the forefoot and small defects. However, because medial sural artery is considered as an important collateral around the knee, the MSAP flap should be avoided in order to prevent the further limb damage in patient with occlusion of popliteal artery [26, 27].