RESULTS
After review of a database of over 200 scapula free tissue transfer reconstructions conducted by the senior author (JMS) at our institution, 17 patients (13 male, 4 female) met criteria to be included in the study (Table 1). All patients were left with total laryngopharyngectomy defects after oncologic ablation necessitating free tissue transfer reconstruction. Average age of the cohort at the time of surgery was 64.5 (SD 9.1, range 46.4-76.3). There were 13 Caucasian patients and the remaining four were African American. Of all patients, 88.2% were primary chemoradiation (n=9) or radiation-alone (n=6) failures undergoing salvage surgery. Only two patients had not received any prior treatment.
All patients had a preoperative diagnosis of head and neck squamous cell carcinoma—the majority of patients had tumors involving the glottis (n=6) and surpraglottis (n=5) with the remaining tumors being of the oropharynx(n=2), nasopharynx (n=2), hypopharynx (n=1), or oral cavity (n=1). Mean follow-up after surgery was 19.1 months with all patients followed for at least 6 months.
Operative notes dictated by the primary operative surgeon (JMS) of all reconstructions were reviewed. All patients underwent microvascular reconstruction with a fasciocutaneous scapular free tissue transfer in a circumferential/tubed (n=11) or non-circumferential, partially-tubed (n=6) fashion. All patients were positioned in lateral decubitus position for flap harvest (Figure 1), and donor sites were closed primarily in all cases (Figure 2). Flap pedicles were based off the subscapular or circumflex scapular artery and veins in all cases. Recipient arteries varied based off availability of healthiest vessels in the neck, specific defect location, and pedicle length, but the most commonly used were the facial (n=5), superior thyroid (n=4) and transverse cervical (n=6) arteries. In regard to veins, the external jugular vein (n=7) or branches off the internal jugular vein (n=9) were used in all cases. An implantable doppler was coupled to the recipient vein in all-cases for continuous monitoring capability in the postoperative period. Mean skin paddle size was 152.2cm2 (SD 56.2cm2, range 67.5-242cm2) and average ischemia time was approximately 4 hours (range 2:57-4:50). It should be noted, that the senior author prefers to complete a majority of the hypopharyngeal reconstruction before starting on construction of the microvascular anastomosis. This results in long ischemia times, which would be modified by starting the microvascular component of the case immediately after repositioning the patient. Salivary bypass tubes were placed in the vast majority of cases at the time of surgery (n=14). There were no significant documented intra-operative complications.
Post-operatively, all patients were admitted to the surgical-trauma ICU for every-1-hour nursing flap checks. Within the patient cohort, there were no partial or total flap losses, resulting in a flap survival rate of 100%. Donor site morbidity was also excellent, with only two post-operative back hematoma requiring bedside drainage and placement of a wound vac. There were no donor site wound infections or dehiscences. Further surgical complications post-operatively included two hematomas (one neck, one chest) and two cases of wound dehiscence (one at flap edge, one at stoma).
Two patients developed pharyngocutaneous fistulas as inpatients (11.7%). One fistula was closed with pectoralis flap successfully, and the other closed after aggressive packing and the use of a wound vacuum device. There was one major medical complication (pulmonary embolism) and 7 minor complications including pneumonia, minor respiratory events, arrhythmias, acute blood loss anemia, delirium, and electrolyte disturbances (Table 2). Mean hospital length of stay was 15.7 days (SD 8.2, range 8-36 days). Adjuvant therapy (chemoradiation, chemotherapy- or radiation-alone) was carried out in a 10 of the 17 patients post-operatively.
Speech outcomes in the outpatient setting were reviewed (Table 3). Pre-operatively, nutritional status was generally poor with only 7 patients able to fulfill all nutritional needs via oral intake. The remaining patients were either supplementing with PEG feeds (n=8) or PEG-dependent (n=2). In the postoperative setting, only one patient remained PEG-dependent, 11 patients continued to supplement their oral intake with PEG feeds, and 5 patients were taking solely by mouth at last documented follow-up. Four patients required esophageal dilation in the operating room as part of their swallowing rehabilitation. In many cases, swallowing outcomes were affected by pre-operative swallowing dysfunction including trismus and multi-level swallowing difficulty. Moreover, swallowing outcomes often worsened after adjuvant treatment with chemotherapy, radiation therapy, or both.
Voice outcomes varied significantly among patients in the cohort (Table 3). Four patients remained with only text/written speech for communication, 6 patients pursued TEP placement and were progressing with voice rehabilitation, and 7 patients were utilizing an electrolarynx with varying degrees of success.