Clinical parameters in evaluating hospitalized patients with orofacial odontogenic infection--a preliminary retrospective study

OBJECTIVES: The purpose of the study was to characterize local and systemic clinical parameters of odontogenic infections in adults, and to isolate parameters that can predict hospital length of stay (LOS). METHODS: File charts of 142 patients, hospitalized between 1994 and 2004 in the Department of Oral and Maxillofacial Surgery of the Sheba Medical Center, were reviewed. Pre-hospitalization data (age, gender, clinical characteristics of the infection, antibiotic intake), and data during hospitalization (blood count, body temperature, heart rate, blood pressure, thrombocytes count, anatomic space involved, cause of infection, type of infection abscess or cellulitis, surgical treatment applied and length of stay) were collected and statistically analyzed to isolate which mostly correlated to LOS--long (4 or more days) or short (less than 4 days). Stepwise logistic regression model was carried out to obtain a predictive model for short or long LOS using the statistically significant parameters. RESULTS: Systemic clinical findings, such as body temperature, blood pressure, heart rate, and WBC count, had normal to slightly elevated values. High blood count PMN, surgical intervention, infraorbital space involvement, and infection in the mandible were significant to LOS. The most important parameter, masking all other significant parameters, was the location of infection. When the maxilla was compared to the mandible, the odds for a long LOS were 3.2 times greater in the mandible. CONCLUSIONS: Orofacial odontogenic infections may be regarded as local infection rather than systemic disease. Hospital LOS is influenced directly by infection location, i.e., lower face (mandible) infection requires a longer LOS.