Functional Outcome
Oral incompetence, ectropion, and trismus occurred in 9 (19.1%), 8 (17.0%), and 6 patients (12.8%), respectively. Again, the extent of resection had no significant impact on the development of any functional impairment (Table III). Nonetheless, solely patients with malignancies originating of the buccal mucosa and oral cavity suffered from oral incompetence (22.7% and 66.7%; p=0.020). Performance of mandibulectomy (p=0.003), but not maxillectomy (p=0.064), affected significantly oral competence, which was otherwise significantly associated with the occurrence of trismus (p=0.009). In addition, size of used free flaps was 14.7 ± 5.1 cm in patients with oral incompetence, which was significantly larger compared to 7.4 ± 2.7 cm in patients with oral competence (p=0.008), while primary tumor size did not significantly differ (4.4 ± 1.4 cm vs. 3.5 ± 1.9; p=0.206) (Table III).
DISCUSSION
We have analyzed clinical outcome of 47 patients with solid malignancies of the cheek that underwent radical tumor resection with creation of partial or through-and-through defects and free flap reconstruction. Within our study, SCCs were the predominant histologic subtype (80.9%) and malignancies mainly originated from the oral cavity (80.8%). This is consistent with literature, reporting mostly on oral carcinomas and rarely on skin carcinomas, requiring cheek reconstruction with free flaps following oncological resections. 3,4Nonetheless, our data further display the great diversity of tumors affecting the cheek region that may hamper analysis of more homogenous subgroups with large patient numbers.
In solid malignancies, and for patients with SCCs in particular, surgical tumor resection with adjuvant therapy in selected cases represents the most frequent treatment modality.5,6,10However, despite radical surgical resection, recurrence rates range from 45.0% to 80.0% in patients with buccal SCCs.11,12Several authors assume that the absence of “real” anatomic boundaries limiting tumor growth and spread might contribute to the high rate of recurrences.11,13 This prompted Ren ZH and coworkers (2017) to perform a more extensive resection of functional anatomic buccal units to achieve better oncologic outcome.3 In fact, they analyzed data of 127 patients with buccal SCCs reporting on significantly better 2-year overall survival (OS: 83.3% vs. 60.1%) and DFS (76.6% vs. 51.9%) in patients undergoing more extensive unit resection compared to conventional surgery.3
Although oncologic principles must supersede reconstructive desires7, we were particularly interested in knowing how the extent of resection impacts functional outcome. Oral incompetence represented the main functional complication occurring in 9 (19.1%) patients followed by occurrence of ectropion and trismus in 8 (17.0%), and 6 patients (12.8%), respectively. This is in line with the results of other publications, reporting on problems with oral incompetence in 4.8% up to 40% of patients with cheek carcinomas.8,14,15 It is noteworthy to mention that the extent of resection had no significant impact on functional outcome in our cohort, which is in accordance to the work of Ren ZH et al. (2017). The authors assume that insignificantly changed functional outcomes in patients with conventional surgery compared to more extensive unit resections, have resulted from the loss of function of preserved structures secondary to induction of fibrosis and loss of functional adjacent structures / attachments by tumor resection.3 However, we found a strong association between oral incompetence and trismus in patients after mandibulectomy. This indicates that functional outcomes more likely depend on the preservation of certain anatomic structures and chosen surgical approach than on the depth of defect.
Recipient site complication rate was 55.3% (n=26), of which wound dehiscence was the most common complication occurring in 29.8% (n=14) of cases. Although the majority of complications were of minor concern, 10 patients developed salivary fistulae (21.3%) that occurred significantly more often in suborbital zone I defects (p=0.035). This is in accordance to former studies reporting on fistula rates of 4.3% to 27.3% of patients.8,14,16-18 The development of fistulae is characteristic for maxillary reconstruction and occurs typically near to the medial canthus (zone I) due to breakdown of suture lines.16 In alignment to that, we observed the highest rate of wound dehiscence (38.5%) in patients with zone I defects compared to 18.2% and 30.4% in zone II and III defects, respectively.
Until now, a number of different free flaps have been described for cheek reconstruction including the ALT 19, RFFF19, FFF 20 and the scapular / parascapular free flap 21. Recently, the versatility of the SAFF has been demonstrated for general head and neck reconstruction, and in particular for cheek and tongue reconstruction.22,23 In our study, the RFFF was used in 31.9% (n=15) of cases followed by the ALT free flap and the scapular / parascapular free flap in 27.7% (n=13) and 21.3% (n=10) of patients, respectively. Among those, the RFFF was mainly harvested for cutaneous reconstruction (73.7%), the ALT flap for reconstruction of myocutaneous defects (83.3%), and the scapular / parascapular free flap for bone reconstruction (62.5%). Our flap survival rate was 95.7%, which is comparable to 95% to 96% reported in former studies.17,19
We believe that the strength of this study lies in its multicentric nature and the analysis of functional outcomes as well as complications. We see three limiting factors: first, the retrospective study design bears an inherent risk of information and selection bias. Second, the heterogeneity of our cohort with solid malignancies originating from different parts of the cheek allows only limited conclusions. Third, the lack of standard measures for functional outcomes in head and neck oncology24 and the fact that functional and aesthetic outcomes have been rated by treating head and neck surgeons as opposed to patient reported outcomes, represent further limitations.