Diffuse deep neck space infection (DDNSI) is an emergent condition that requires efficient surgical and airway management. A novel surgical approach has been developed to enhance access and improve visualization of the involved deep neck spaces, which allows for better evaluation of the extent of tissue necrosis. This study will compare the treatment of DDNSI with traditional incision and drainage with drain placement (Conventional) versus the new incision and drainage design with the use of a wound vacuum (VAC). This retrospective cohort study was performed on DDNSI cases treated with the VAC versus Conventional techniques from July 2014 to September 2020 at Memorial Hermann Hospital by the oral and maxillofacial surgery service. To be categorized as a DDNSI, the patient had to demonstrate radiographic evidence of an infection bilaterally in a minimum of four spaces, such as bilateral submandibular, sublingual, and submental spaces. Primary predictor variable was treatment method for DDNSI, Conventional versus VAC. Primary outcome variables were hospital length of stay (LOS), number of washouts, and days of intubation. Covariates were age, sex, number of spaces involved, presence of necrotic tissues, comorbidities, and mortality. Appropriate uni- and bi-variate statistics were calculated. Statistical significance was set a P < .05. Fifty-one patients (17 female and 34 male) aged 18 to 65years were treated for DDNSI. Twenty-eight patients were treated using the VAC approach and 23 patients were treated with the conventional approach. The average LOS 8.3± 0.8days (P-value=.0001), number of days intubated 3± 0.3 (P-value=.0001), and number of required washouts 2± 0.2 (P-value=.004) were statistically lower in the VAC group compared to the Conventional group. There were significant improvements encountered with overall length of hospital stay, number of days intubated, and the number of required wash outs. There were significant differences in outcomes between patients with comorbidities when compared to those who had none.
Read full abstract