(23) Seasonal Variability in Odontogenic Infections Requiring Surgical Intervention in the Operating Room: A Retrospective Cohort Analysis

Abstract: The incidence of odontogenic infections requiring surgical intervention may vary throughout the year, potentially following a seasonal pattern. The purpose of this study is to determine whether there is a quantifiable and statistically significant yearly periodicity in the volume of operative odontogenic infections. We hypothesized the summer months would exhibit the highest volume due to environmental and behavioral factors, such as changes in outdoor activity and healthcare access.
Materials and

Methods: To assess seasonal variation in odontogenic infections, we conducted a retrospective cohort study. The Louisiana State University, New Orleans, Oral and Maxillofacial Surgery surgical log was analyzed by Current Procedural Terminology (CPT) codes for incision and drain. Monthly CPT code volumes were analyzed as the dependent variable, with the calendar month as the predictor variable to assess seasonal variation. Monthly CPT volume was collected over an 82-month period from 2018 to 2024. The raw data were visualized as a time series and adjusted using a ratio to the moving average. Autocorrelation analysis was performed on the transformed dataset to identify yearly periodicity. Multivariable modeling quantified the proportion of volume variability (R2) attributed to seasonal trends. A sub analysis evaluated the presence and strength of periodicity across four age groups.

Results: A total of 789 CPT codes obtained from the surgical logs were included in the study. Mean age of the cohort was 40.65 with standard deviation 16.4 and range of 93. June and July had the highest volume of odontogenic infections undergoing operative room intervention and January and April exhibited the lowest volume. Autocorrelation analysis of the transformed 82-month dataset did not show statistically significant positive or negative peaks with a lag from 2 to 15 months, indicating there was no significant periodicity. Multivariable modeling of monthly patient counts, adjusted using the ratio to the moving average, showed no statistically significant (P values > .34) association with monthly explanatory dummy variables across the 4 age groups in the sub-analysis: R2 = 0.1384 ages 0-17, R2 = 0.3258 ages 18-44, R2 = 0.11 ages 45-64, and R2 = 0.18 ages > 65.

Conclusion: This study aimed to determine whether odontogenic infections requiring surgical intervention follow a quantifiable seasonal pattern. Although fluctuations in monthly surgical volumes were observed, with peaks in June and July and lower volumes in January and April, statistical analysis did not demonstrate significant yearly periodicity. Autocorrelation analysis and multivariable modeling failed to identify a consistent seasonal trend across age groups. These findings indicate that while operative volume varies throughout the year, these changes are likely influenced by factors beyond predictable seasonal patterns. Given the absence of a statistically significant trend, fluctuations in case volume may be insufficient to guide staffing or operating room scheduling adjustments. Surgeons should prioritize real-time case monitoring over seasonal forecasting for resource allocation. Further research is warranted to explore other potential determinants, such as access to care and environmental influences, that may impact infection trends.

References:
1. Yankov, Yanko G et al. “Etiology and Demographic Distribution of Odontogenic Abscesses in the Maxillofacial Area in Patients Over 18 Years of Age: A Five-Year Retrospective Study.” Cureus vol. 16,4 e59334. 30 Apr. 2024, doi:10.7759/cureus.59334
2. Lixandru, Cosmin Ionuţ et al. “Oral infections - a retrospective study of patients treated in the Oral and Maxillofacial Surgery clinic of the Emergency County Clinical Hospital in Sibiu.” Medicine and pharmacy reports vol. 97,3 (2024): 380-389. doi:10.15386/mpr-2759
Average monthly patient count over 82 monthsTime series plot depicting monthly patient count over 82 months and the moving averageAutocorrelation analysis to the transformed 82-month dataset