Abstract
This study analyzed the surgical outcome and complications of 1000 microvascular free flaps performed at the authors’ institution in Germany, between 1987 and 2010. 972 patients underwent reconstruction with 1000 flaps: 28% latissimus dorsi flaps, 27% radial forearm flaps, 20% iliac crest flaps, 12% fibula flaps, 6% jejunal flaps, 2% anterolateral thigh flaps, and 5% other flaps. 130 failures (7.6%) were encountered, including 58 complete flap failures (44.6%) and 72 partial free-flap failures (55.4%). This study confirms that free flaps are extremely reliable in achieving successful reconstruction of the head and neck, but it is essential that complications be recognized and addressed early in their course to prevent or minimize devastating consequences. Owing to the large number of possible errors in flap transplantation, microsurgeons should always check everything for themselves. The on-duty doctors and nursing staff should not be trusted blindly. Venous thrombosis and cervical haematoma are the most common complications at the recipient site and are mainly responsible for flap failure, while complications occurring at the donor site may result from dehiscence and graft necrosis. When a compromised flap is identified, surgical re-exploration should not be deferred.
Microvascular free tissue transfer has gained world-wide acceptance as a means of reconstructing post-oncologic surgical defects in the head and neck region. The use of microvascular free tissue transfer has allowed the reconstruction of complex defects in patients after head and neck cancer surgery. The development and refinement of microvascular instruments and magnification have improved the overall success rates reported to 94–96%; but uncommon complications may prove devastating. Free tissue transfer has become a useful procedure that provides a great advance for head and neck reconstructions, while the standardized techniques have improved the results.
Microvascular free flap transfer can be performed safely in the elderly, a growing segment of society, offering an improvement in quality of life despite the advanced age of the patients. Medical complications and overall length of hospital stay are higher in the elderly population, reflecting their increased associated co-morbidity. To balance the risks and benefits of newly described modifications against the results obtained with established methods, a critical review of the recent results and their comparison with series published by other centres is of great value.
In microsurgery, it is essential to identify and manage complications as early as possible in an effort to preserve or salvage the viability of the flap. This study evaluates the outcome and complications of 1000 microvascular free flaps performed at the Department of Oral and Maxillofacial Surgery of the University Medical Centre Hamburg-Eppendorf from 1987 to 2010. Complications of microvascular surgery have been classified into recipient site or donor site complications. The systemic complications were also analyzed.
Material and methods
The records of 972 consecutive patients in whom 1000 microvascular free flap procedures for reconstruction of defects in the head and neck region were performed during the years 1987–2010 were reviewed retrospectively. The patients ranged in age from 4 to 92 years, with a mean age of 56.6 years, and consisted of 632 males and 340 females. The medical records and the details of the primary tumour site, flap type, outcome and complications were analyzed. The free flaps selected for reconstruction are listed in Fig. 1 . 72% of the defects arose after the treatment of malignant neoplasm of the oral cavity. 16% of the defects resulted from the resection of mid or upper face and skull-base tumours and another 12% from treatment of hypopharyngo-esophageal tumours. 33% of reconstructions were carried out in patients after radiotherapy. Primary reconstructions were performed in cases with benign diagnoses (62 patients), whereas secondary reconstruction was the therapeutic strategy in malignant diseases (832 patients). The two-stage approach was used to shorten the operation time, placing the patient under less stress. Compared with the limitations of frozen-section examinations, particularly in difficult histopathologic examinations (lymph nodes and soft-tissue and bone tumours), conventional paraffin sections provided advantages in terms of diagnostic certainty. A final histologic diagnosis is useful, in particular before complex reconstructions. Reconstructions not related directly to tumour reconstruction, such as therapy for osteoradionecrosis, were conducted in 78 patients.
The complications after microsurgery were divided into major and minor complications. A major flap or donor-site complication was one that required additional surgery. A major systemic complication was defined as potentially life-threatening. Minor complications were those that resolved on their own or required minimal intervention.
Results
In this clinical series, the primary site of the tumour was the oral cavity in 699 patients (72%), the mid or upper face and skull base in 155 patients (16%), and the hypopharynx or oesophagus in 116 patients (12%). 972 patients had reconstructions with 1000 flaps: 28% were latissimus dorsi flaps, 27% were radial forearm flaps, 20% were iliac crest flaps, 12% were fibula flaps, 6% were jejunal flaps, 2% were anterolateral thigh flaps (performed since 2009), and 5% were other flaps ( Fig. 1 ).
Complications at the recipient site
There were 130 failures (7.6%). At the recipient site 58 total flap losses (5.8%) and 29 cases of major bleeding (2.9%) were identified ( Fig. 2 ). 121 free flaps required surgical revision and the success rate of revisional surgery was 48%. Vascular occlusion (thrombosis) of one of the vessels was the primary (4.5% arterial, 6.8% venous) reason for flap loss, with venous thrombosis being more common than arterial occlusion. The majority of flap failures occur within the first 36 h. The main reasons for flap failure were the use of compression bandages, tapes around the neck or tracheal strips with extrinsic compression of the vascular pedicle ( n = 7). Tight wound closure or wound haematoma may also compromise the flap by obstructing venous outflow. Major bleeding (4.9%) was a common cause of failure in microvascular free tissue transfer.
Minor complications included 72 partial flap failures (7.2%) and 57 cervical hematomas (5.7%). The management of total free flap failures was based on subsequent free flap transfer ( n = 59), on a regional flap, or on secondary healing. New anastomoses of the vessels had to be performed in 44 (4%) cases in the early postoperative period. 34 flaps could be salvaged.
The most reliable flap regarding survival was the radial forearm flap (total flap failure in 3.3% of the cases). The latissimus dorsi flap and the iliac crest flap failed in 8.4% and 12% of cases. The length of intensive care unit stay was 3.8 days, and the total hospital stay was 36 days.
Complications at the donor site
Major donor site complications were observed in 80 cases (8%). The most common were wound dehiscence, tendinous exposition and graft necrosis ( Fig. 3 ). In particular, covering the radial forearm defect with full thickness skin led to problems because of the exposed tendons. The removal of the latissimus dorsi flap led to the fewest problems. Even larger defects could be covered relatively easily primarily, as well as defects in the lateral thigh after removal of anterolateral thigh (ALT) flaps (with a width of less than 12 cm).
Systemic medical complications
Systemic medical complications developed in 90 of the 972 patients (9.3%), including 35 minor complications and 55 major complications. The most common major complication was respiratory insufficiency ( n = 16). A patient with oral floor carcinoma developed half-side paralysis after transplantation of a microsurgical latissimus dorsi flap. Major complications involved 8 deaths (0.8%). Minor systemic complications occurred in 3.5% of the cases, including pneumonia in 2.2% and hypertensive crisis in 0.7% ( Fig. 4 ).
Results
In this clinical series, the primary site of the tumour was the oral cavity in 699 patients (72%), the mid or upper face and skull base in 155 patients (16%), and the hypopharynx or oesophagus in 116 patients (12%). 972 patients had reconstructions with 1000 flaps: 28% were latissimus dorsi flaps, 27% were radial forearm flaps, 20% were iliac crest flaps, 12% were fibula flaps, 6% were jejunal flaps, 2% were anterolateral thigh flaps (performed since 2009), and 5% were other flaps ( Fig. 1 ).
Complications at the recipient site
There were 130 failures (7.6%). At the recipient site 58 total flap losses (5.8%) and 29 cases of major bleeding (2.9%) were identified ( Fig. 2 ). 121 free flaps required surgical revision and the success rate of revisional surgery was 48%. Vascular occlusion (thrombosis) of one of the vessels was the primary (4.5% arterial, 6.8% venous) reason for flap loss, with venous thrombosis being more common than arterial occlusion. The majority of flap failures occur within the first 36 h. The main reasons for flap failure were the use of compression bandages, tapes around the neck or tracheal strips with extrinsic compression of the vascular pedicle ( n = 7). Tight wound closure or wound haematoma may also compromise the flap by obstructing venous outflow. Major bleeding (4.9%) was a common cause of failure in microvascular free tissue transfer.
