Adverse Outcomes Following Free Tissue Transfer: A 3-year Experience From a Major Referral Centre

Free tissue transfer encompasses a variety of techniques by which tissue is moved to another region of the body, with anastomosis of the divided artery and vein. Currently, success rates are reported at 91-99% [1], however, little is known regarding predictors for adverse outcomes.We aim at identifying predictors for negative outcomes following free ap surgery; and predict that elderly patients and patients with head and neck free aps will have inferior outcomes due to comorbidity. Methods


Introduction
The rst recorded successful free ap reconstruction was reported in 1972 [2]. Since this time, numerous techniques have evolved in relation to the transfer of free tissue from one region of the body to another. Microvascular surgery is utilised to anastomose donor blood vessels to regional blood supply at the recipient site. Free tissue transfer is the nal rung in the reconstructive ladder and should only be utilised when other all reconstructive techniques are deemed inferior for the clinical problem at hand. For this reason, thought must be given to potential perioperative complications that can adversely affect patient outcomes.
Since 1972, surgical techniques pertaining to free tissue transfer have continued to improve. Currently, success rates are reported between 91 to 99%. [1] Regardless, this patient population is at great risk for postoperative complications. Formeister reported that 45.6% of patients undergoing free tissue transfer experienced an adverse event during admission. [3] Tam found that 18.8% of patients required an unplanned return to theatre. [4] For free bular bone aps, it has been reported that 48 major recipient site complications occurred in 41 reconstructions. [5] Because free ap patients are at signi cant risk of adverse outcomes, it is important to identify predictors for complications and to identify groups most at risk.
Previous publications have found patient risk factors for complications, however, to date, there is no clear consensus regarding free tissue transfers. Low skeletal muscle mass [6], increased alcohol consumption [4], age [3,7,8], higher ASA [8], ap type, and site of reconstruction [3] have all been shown to increase the incidence of medical and surgical complications. Obesity has not demonstrated an increased number of adverse events. In fact, higher body weight has been protective in some cases [9]. Many authors have been unable to nd an association between advanced age and harmful outcomes [10][11][12][13][14]. Further research is needed to delineate associations between free ap reconstructive surgery and any subsequent complications.
The objective of this study is to identify predictors for negative outcomes following free ap surgery. We aim to compare free aps by body region, age, and gender. We predict that elderly patients and patients with head and neck free aps will have inferior outcomes because these populations are more likely affected by co-morbidity.

Methods
This is a single centre retrospective case series. All cases of free tissue transfer occurring between 28th February 2018 and 17th Feburary 2021 were collated using the theatre electronic medical records system at a large tertiary centre in Adelaide, South Australia. 67 cases were identi ed. Case les and discharge summaries were then retrospectively analysed to identify complications that occurred as a result of the operation. Chi squared hypothesis testing was used to identify factors that contribute to negative outcomes for patients. Patients outcomes were deliniated by body region (head and neck, breast, limbs), age, and gender.
Ethics approval was sought from the Southern Adelaide Clinical Human Research Ethics Committee and the need for informed consent was waived due to the retrospective nature of the study. Research was performed in accordance with relevant guidelines and regulations. Patient data was stored and managed in accordance with the Australian Code for the responsible Conduct of research. Informed consent is not required for this type of study. The authors have no con ict of interest to disclose. The authors did not receive support from any organisation for the submitted work. The authors received no nancial support for the research, authorship, and/or publication of this article. Table 1 provides an overview of the patient demographic data included in this study. A total of 67 patients underwent free tissue transfer. Patients ranged from 18 to 86 years of age, with an average of 57. Almost half of the patients were aged between 46 to 64 years. There were 37 males and 30 females, with a ratio of 1.23 to 1. 46.3% of patients in this cohort experienced at least one complication. This was more likely to occur in the elderly with 61.1% of patients over the age of 65 being affected. Only 40.8% of patients younger than 65 experienced a complication. This trend also holds true for major complications.

Patient Demographics
The elderly were more than twice as likely to experience a major problem (de ned as a Clavien-Dindo Classi cation grade of IIIa or higher).
[15] 44.4% of cases required substantial unplanned post-operative intervention, compared to 20.4% of the younger cohort. Breast reconstruction patients were less likely to experience any complication with 36.8% of cases affected, compared to 48.4% of head and neck patients and 52.9% of limb patients. No differences were found in terms of incidence of major complications between body regions. Complications by age Table 2 categorises adverse events data by age. The accompanying gure 1 provides a visual analysis of the same. A total of 49 patients were younger than 65 and 18 patients were older than 65. It was found that older patients were more likely to require an unplanned return to theatre and were much more likely to develop a post-operative infection of the reconstructed site. The odds ratio for a patient over the age of 65 to require a return to theatre was 4.1. This was statistically signi cant with a p-value of 0.017. The odds ratio for developing a free ap infection was 13.71. This was statistically signi cant with a p-value of 0.0053. No signi cant difference was found in terms of death, unplanned ICU, anaemia requiring transfusion, and rates of total ap failure.  Table 3 groups complications data by the site of surgery. This information is also presented visually in gure 2. A total of 31 patients underwent head and neck free aps, 19 patients underwent breast free aps, and 17 patients underwent limb free aps. It was found that free ap location was signi cant in determining whether patients would require a subsequent blood transfusion, with a p-value of 0.0071.
There was no statistically signi cant difference in terms of complication rates between body regions for death, unplanned ICU admission, unplanned return to theatre, ap failure, and reconstructed site infection. There were no statistically signi cant differences found between males and females in terms of postoperative free ap complication rates.

Discussion
This study demonstrates that free aps are a generally safe and successful procedure. The overall mortality was 1.5% and overall ap success was 92.5%. These numbers are similar to other studies [1,16]. There was no difference in incidence of total ap failure or death in patients older than 65 years. It can therefore be concluded that age does not predispose to failure of the procedure. In keeping with our ndings, Ferrari et al concludes that as long as patients are carefully selected pre-operatively in terms of comorbidities, age should not be a disqualifying factor for reconstruction with free aps. [16] The total percentage of cases affected by a complication according to the Clavien-Dindo classi cation was 46.3%. This is in line with other publications. [3] The likelihood of a major complication was 26.9%.
The elderly were more likely to experience any complication and much more likely to experience a grade IIIa complication or higher, compared to younger patients. Breast patients were less likely to experience any complication, compared to other body regions. No difference was found in terms of major complications between body regions. It can therefore be concluded that age is a suitable predictor of worse post-operative outcomes, while breast reconstruction patients are more likely to have a favourable recovery period.
Overall, it was found that the elderly were much more likely to require reoperation and develop postoperative reconstructed site infection. 45% of free aps over the age of 65 required an unplanned return to theatre. The most common reasons for this were wound breakdown, ap congestion, and haematoma formation. 22% of elderly patients also developed infection of the reconstructed site and required treatment with antibiotics. Only 2% of younger patients also experienced this complication. Currently, the literature suggests that elderly patients are 3-4 times more likely to develop chronic wounds because of di culty with wound healing. This is thought to be secondary to prolongation of the in ammatory phase of healing, macrophage dysfunction, and malnutrition. [17] Regardless, these ndings highlight the need for careful monitoring of elderly patients post operatively, in terms of ap health, but also in terms of overall physiology. There should be a low threshold to initiate antibiotic prophylaxis when free ap infection is suspected in patients aged 65 or older.
Head and neck free aps are most commonly performed following resection of cancers in this body region. In general, these patients are well known to be more medically co-morbid than other patient populations. In this patient population, the average age for patients with head and neck free aps was 64. This was 12 years older, on average, than breast patients and 14 years older than limb patients. Since elderly patients are more likely to be co-morbid, this patient population would have therefore been considered a higher pre-operative risk. Our ndings suggest that despite this, the rate of all complications is not signi cantly increased. Head and neck free aps are not more likely to result in complications.
Other authors report similar results, although it is interesting to note that Las et al found signi cantly reduced rates of post-operative infection and total ap failure in patients undergoing breast reconstruction. [18] In this study, patients undergoing limb operations were found to be signi cantly more likely to require a blood transfusion peri-operatively. The most common indication for reconstruction involving limbs was trauma. Many of these free aps were performed following orthopaedic intervention or to reconstruct signi cant soft tissue loss following trauma. Blood loss is therefore more likely related to the mechanism of injury and orthopaedic interventions, rather than directly to the free ap itself. Regardless, these ndings suggest that patients undergoing limb free ap reconstruction, especially in the setting of trauma, will require more careful monitoring of haemoglobin levels than other patient populations.
As predicted, there was no difference in complication incidence between male and female patients.
The main limitation of this study is that this is a single centre retrospective analysis. Generalisability is therefore limited. The sample size is also small.

Conclusion
This study found that elderly patients were more likely to experience any complication and much more likely to experience a major complication. In particular, patients aged 65 and older were more likely to require an unplanned return to theatre (OR 4.1) and were much more likely to develop a post-operative reconstructed site infection (OR 13.7). This highlights the need for careful post-operative monitoring of elderly patients. Patients undergoing limb operations were more likely to require a blood transfusion, although this is likely related to trauma as the most common indication for surgery. No difference was found in terms of complication rates between head and neck patients and other patient groups.
Declarations DISCLOSURES Ethics approval was sought from the Southern Adelaide Clinical Human Research Ethics Committee and the need for informed consent was waived due to the retrospective nature of the study. Informed consent is not required for this type of study. The authors have no con ict of interest to disclose. The authors did not receive support from any organisation for the submitted work. The authors received no nancial support for the research, authorship, and/or publication of this article.  Complications by body region