(99) Incidence of Complications Associated With Non-Invasive Positive Pressure Ventilation in Maxillofacial Trauma Patients

Abstract: Maxillofacial trauma patients with paranasal sinus or orbital fractures are routinely placed on sinus precautions. The basis is the assumption that sinus pressure can displace sinus flora into the surrounding soft tissues resulting in complications such as cellulitis, infection, or emphysema. Although intuitive, the necessity of these precautions remains unproven with limited supporting literature; There are two case reports of patients with orbital fractures who developed orbital emphysema after nose blowing and sneezing.^1,2 Non-invasive positive pressure ventilation (NPPV) is a standard treatment for acute respiratory failure or obstructive sleep apnea, but its use in patients with facial fractures remains controversial. This study aims to evaluate whether NPPV use in maxillofacial trauma patients increases the incidence of maxillofacial cellulitis, infection, emphysema, or orbital complications. A retrospective cohort study was conducted using the University of Cincinnati Medical Center (UCMC) trauma patient registry from 2010 to 2023. Eligible patients were over the age of 18 with paranasal sinus or orbital fractures. The primary predictor variable was use of NPPV. The primary outcomes were maxillofacial infection, cellulitis, and emphysema, as well as orbital emphysema (OE) and orbital compartment syndrome (OCS). Secondary outcomes included loss of visual acuity and procedures required for treatment of infection, emphysema, and OCS. Covariates included were age, sex, race, ethnicity, alive/dead, cause of injury, fracture type, injury location, length of stay, and time to procedure. Cross-tabulations compared variable characteristics between the NPPV and non-NPPV groups (Chi-Square, Fisher’s Exact, T-Test). Logistic regression models were used to determine the effect of NPPV on primary and secondary outcomes, adjusting for confounders. Per the protocol the study sample was composed of 15 subjects which received NPPV, and 15 who did not. For each group, subjects were randomly selected from the population that met eligibility criteria. The study sample was made of 19 males (63%) and 11 females (37%). The sample was 87% White, 13% non-White, and 100% non-Hispanic. All subject outcomes were alive. The mean sample age was 57 (standard deviation (SD) ± 23). The average length of hospital stay was 3 days (SD ± 4), and the average time to procedure was 19 days (SD ± 23). The only covariate which showed a statistically significant difference between study and control groups was time to procedure (P = .013). Causes of injury, race (White), fracture type, and injury location showed a statistically significant relationship to developing OCS (P < .001). Fractures of inferior orbital wall and maxilla showed a statistically significant relationship to developing OE (P < .001). Primary outcomes of maxillofacial infection, cellulitis, and emphysema were not observed in any subjects. OE developed in 3 subjects who did not receive NPPV, and no subjects who received NPPV. OCS occurred in 1 patient who received NPPV, and did not occur in patients who did not receive NPPV. Neither OE nor OCS demonstrated a statistically significant association with the predictor variable (P > .2). Secondary outcome loss of visual acuity was present in 4 patients (13%), all of whom received PEEP. There was no statistically significant association with the predictor variable (P = .7). No other secondary outcomes were observed in any subjects. In conclusion, this study suggests that NPPV use in maxillofacial trauma patients with paranasal sinus or orbital fractures does not significantly increase the risk of OE, or OCS, or loss of visual acuity.

References:
1. Dong, Q. N., et al. "Retrobulbar orbital emphysema associated with medial orbital wall fracture." Journal of Craniofacial Surgery 30.5 (2019): 1549-1551.
2. Clopper, C. J., & Pearson, E. S. "The use of confidence or fiducial limits illustrated in the case of the binomial." Biometrika 26 (1934): 404-413.