Oral squamous cell carcinoma (SCC) is the most common malignancy among head and neck cancers; it is characterized by local invasion and lymph node metastasis. Most of these tumors present in advanced stages at initial diagnosis (1, 2). Surgery is the preferred therapeutic approach, and defects caused by surgical excision are usually repaired using free flaps (3).

Complications after free flap reconstruction are frequently observed (4). An example would be an orocutaneous fistula (OCF). It can significantly delay the initiation of adjuvant therapies, increase the duration of a patient’s admission and costs of treatment, delay oral feeding and voice rehabilitation, and increase the risk of carotid blowout (5–7). Several prophylaxis and management procedures have been introduced. These consist of conventional wound care, negative pressure wound therapy, and surgical management (8–11), each having its corresponding indications. Moreover, numerous scholars have explored the predictors of fistula development in head and neck surgery. The risk factors comprise age, unhealthy lifestyle, systemic disease, tumor stage, previous radiotherapy, nutrition status, tumor site, and treatment method (12). However, these findings were concluded based on SCC in larynx and hypopharynx, which are significantly different from oral SCC.

To our best knowledge, there are only two studies on the predictors of OCF formation. Dawson et al. (13) enrolled 102 patients with different etiologies, such as cancer, osteoradionecrosis, ameloblastoma, and trauma for head and neck surgery, and found that OCF occurred in 11% of the patients. The only statistically significant association identified was between the formation of an OCF and previous chemoradiotherapy. Girkar et al. (14) reported a rate of 9% based on 587 eligible oral SCC patients, and the presence of surgical site infection and performing bilateral neck dissection posed the maximum risk for developing OCF.

Therefore, this current study aims to identify the risk factors associated with OCF formation after free flap reconstruction for oral SCC.

A total of 856 patients were included, including 590 men and 266 women. The median age was 49 (range: 22–74) years. There were 573 smokers and 377 drinkers. Preoperative low hemoglobin and albumin levels occurred in 204 and 146 patients, respectively. A total of 234, 100, and 88 patients had hypertension, diabetes mellitus, and cardiovascular and cerebrovascular disease, respectively. Cachexia was noted in 378 patients ( Table 1 ).

The distribution of the primary sites was as follows: tongue/floor of the mouth (TFOM) in 455 patients, buccal in 234, and lower gingiva in 167. For the tumor stages, T2 in 279 patients, T3 in 444, and T4 in 133. Unilateral and bilateral neck dissection was performed in 702 and 154 patients, respectively. There were 386 radial free forearm flaps, 300 anterolateral thigh flaps, and 170 free fibula flaps. The muscles in the floor of the mouth (pull-through procedure) were resected in 134 patients.

SSI occurred in 199 patients. Flap re-exploration was required in 90 patients, and 65 were successfully salvaged. Twenty-five flaps were lost, and among these patients, 20 underwent an immediate pectoralis major myocutaneous flap reconstruction, and five received direct suture. The overall success rate of free flap reconstruction was 97.1%.

To our knowledge, this study included the largest sample size for analyzing the risk factors of OCF formation after oral SCC surgery. OCF formation after free flap reconstruction was common. The presence of cachexia, tumor site at the TFOM, T4 stage, and SSI significantly increased the risk of OCF. Although it requires a long period, conventional wound care can obtain satisfactory outcomes in OCF management.

Cachexia is related to weight loss, inadequate food intake, inactivity, loss of muscle mass, and metabolic derangement in patients with advanced cancer (16). It could result in more toxicities, poorer disease-free survival, and more treatment gaps (20). More importantly, postoperative complications are associated with cachexia. Among 2,531 patients undergoing hepatectomy, Lee et al. (21) reportedly had a higher occurrence of bleeding, infection, wound complication, and respiratory failure in the malnutrition cohort. A similar finding was also confirmed in other solid malignancies (22), but there are only a few reports regarding head and neck cancers. Hayashi et al. (23) enrolled 192 patients receiving chemoradiotherapy for head and neck cancer, and found that in a setting of chemoradiotherapy, all adverse events were strongly associated with cachexia. Among grade 3–4 adverse events, the frequency of leukopenia, anemia, and neutropenia had significant differences between patients with and without cachexia. The current study may be the first to identify the profound significance of cachexia in causing OCF formation; cachexia impaired wound healing, causing fistula formation. This finding carries meaningful clinical importance as it indicates that greater attention is required for cachexia patients to prevent OCF development.

The primary site, tumor stage, and pull-through procedure were significantly associated and also predictive factors. Due to the special location of TFOM in the oral cavity, treatment of these advanced-stage tumors often required mandibular lip-split surgery or pull-through resection, and extensive excision usually resulted in the creation of dead space. Al Deek et al. (6) emphasized the importance of dead space obliteration in preventing OCF occurrence. The dead space between the mandible and hyoid bone was originally occupied by extrinsic tongue muscles, and should have been filled with viable tissue during reconstruction. The part of the flap stuffed in the dead space should be checked for reliability after establishing perfusion, especially when the excess muscle was resected or its orientation was changed during the inset of the flap. Our findings also supported this viewpoint. Both primary TFOM site and T4 stage were associated with higher rates of OCF formation. Similarly, Girkar et al. (14) reported OCF development in approximately 15% of patients with TFOM SCC compared to 7% in patients with SCC in other subsites, irrespective of the type of flap used. Additionally, even if early oral intake is not allowed, motion of the reconstructed TFOM still occurs during the recovery period. If these movements occur after a non-ideal design, they could impair the water-tight closure leading to saliva accumulation at the TFOM as this is a dependent part of the oral cavity. Thus, tension must be distributed by multilayer mattress sutures to minimize dehiscence and suture line tearing.

SSI was also a factor that could not be ignored. It occurred in up to 15%-45% of the patients undergoing surgery for head and neck SCC (17, 18, 24). SSI was the result of the interaction of multiple elements, such as prolonged surgical duration, faulty surgical techniques, poor oral hygiene, poor nutritional status, type of reconstruction, and resection range. The negative effects of these factors on incision healing was confirmed in pharyngeal cutaneous fistula (4). Generally, fistula and infections coexist, resulting in a vicious cycle where one may lead to the other. Consistent with the finding of Girkar et al. (14), the current study clarified the effect of SSI on OCF formation.

Other researchers have reported the predictive value of bilateral neck dissection and previous chemoradiotherapy (13, 14). The association between neck dissection and OCF formation may be explained by increased blood loss and increased duration of surgery. Moreover, bilateral neck dissection was only performed if the primary tumors extended across the midline or if there were pathologically proven lymph nodes in the contralateral neck, indicating an extensive disease. However, the current study failed to note this positive relationship. A difference in the study design might have contributed to this variation. Previous chemoradiotherapy significantly posed an apparent effect on wound healing and tissue fibrosis (25). This dysfunction could explain the tendency of OCF formation in patients with previous chemoradiotherapy.

The treatment of OCF is often time-consuming, usually requiring 4 weeks or longer (5). Currently, there are three kinds of management for OCF, conventional wound care, negative pressure wound therapy, and surgical intervention (6–11). Numerous studies have demonstrated the reliability of negative pressure wound therapy. Tian et al. (8) reported successful closure in 90% of OCFs via this method with a median time of 19 days. Inatomi et al. (10) also reported that 82.4% of fistulas were addressed by this method, with a mean closure time of 30.4 days. A recent review by Khoo et al. (5) confirmed the superiority of negative pressure wound therapy over conventional wound care by reducing the closure time, but with a similar success rate. Secondary surgery was usually the last choice after conservative treatment. Our findings also showed that conventional wound care was achieved at a success rate of 86.2% using local flap for the residual cases.

Limitations in the current study must be acknowledged. First, it was retrospective, thus possessing inherent bias. Second, we did not calculate the effect of OCF on oncologic survival. Third, routine lower lip splitting was debatable and may increase additional risk of OCF occurrence.

In summary, OCF formation after free flap reconstruction was common. The presence of cachexia, tumor at the TFOM site, T4 stage, and SSI significantly increased the risk of OCF formation. Although it required a long period, conventional wound care could obtain satisfactory outcomes in OCF management.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The Institutional Research Committee of Henan Cancer Hospital approved our study, and all participants signed an informed consent agreement. The patients/participants provided their written informed consent to participate in this study.

All the authors made a contribution in study design, manuscript writing, study selection, data analysis, study quality evaluation, and manuscript revision. All authors have read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.