Introduction: Current literature on morbidity and mortality in total laryngectomy is both limited and highly variable in scope. Pharyngocutaneous fistula is a major complication of this procedure, which adversely affects patient recovery and may delay needed adjuvant therapy. Fistula and wound complications have been associated with several factors, such as poor nutritional status and prior radiation therapy, and regional or free flap reconstruction is often used to reduce the rate of fistula formation. However, more complete assessment of postoperative complications in this patient group is necessary for effective preoperative planning and postoperative care.
Methods: The Veterans Affairs Surgical Quality Improvement Project (VASQIP) database was queried for patients that underwent total laryngectomy between 1991-2015. Patient demographics and operative information, including age, gender, BMI, medical co-morbidities, pre-operative lab values, and post-operative outcomes including wound complications, death within 30 days, and total complications were obtained and converted to categorical values when applicable. Chi-square and binary logistic regression analyses were conducted with a p value <0.05 indicating statistical significance.
Results: A total of 5429 total laryngectomy cases were identified in the VASQIP database. 361 free flap reconstructions and 933 neck dissections were identified in this cohort, for an overall rate of 6.6% and 17.2% respectively. The overall incidence of wound complications in this cohort was 412, for an overall rate of 7.6%. Free flap reconstructions carried a wound complication rate of 13.3%. Neck dissection demonstrated a higher wound complication rate at 21.8%. On univariate analysis, free flap (p<0.001), and neck dissection (p=0.012) were both predictive of wound dehiscence. Additionally, free flap and neck dissection were predictive of the development of any wound complication (p<0.001, p = 0.003, respectively) while only free flap was associated with a higher risk of death within 30 days (p=0.006). On regression analysis, pre-operative platelet count, male gender, renal failure, free flap, neck dissection, and operation time were significant for total wound complications. Specifically, male gender, free flap reconstruction, preoperative platelet count, and operation time were significant for wound dehiscence.
Conclusions: The cohort analyzed here is one of the largest comparing patients undergoing free flap reconstruction or neck dissection with those who did not for total laryngectomy. Unfortunately, the VASQIP database does not contain specific data on fistula formation, nor does it record the presence of prior radiation therapy. The higher rate of wound complications in the flap reconstruction group likely reflects the use of this reconstruction in previously radiated patients. Interestingly, neck dissection also increased the rate of wound complications, which may reflect the more extensive procedure involved. These data highlight the need for further studies on the optimal extent of surgery in laryngeal cancer patients, considering oncologic, functional, and complication outcomes.