Timing of perioperative oral care and postoperative pneumonia: a propensity score-matched cohort study.

BACKGROUND: Oral care is considered effective for preventing postoperative pneumonia (POP), though optimal timing for its implementation remains unclear. Opinions regarding the timing of oral care in clinical practice are conflicting, as some clinicians consider that reducing the oral bacterial load immediately before intubation is effective, whereas others are concerned that oral care performed just before surgery may promote bacterial dispersion and microaspiration. As a result, there is no consensus regarding the optimal timing of intervention. In addition, given the limited human resources and time available in perioperative care, identifying the timing that maximizes preventive effectiveness is important for optimizing clinical practice. METHODS: This retrospective cohort study examined the associations of presence and timing of oral care with POP incidence following surgery under general anaesthesia. To account for potential confounding factors related to patient background and perioperative severity, propensity score matching and multivariate analyses were performed. POP, the primary outcome, was determined based on ICD-10 codes and initiation of antibiotic administration. Analysis was conducted using multivariate logistic regression. RESULTS: Before propensity score matching, 1,783 patients were included in the analysis, from which 376 matched pairs were selected. Patients who underwent intervention had a lower POP incidence rate (1.6% vs. 4.3%), which was an independent risk-reducing factor (aOR 0.27, 95% CI 0.14–0.52). Furthermore, patients with intervention were divided into three groups (preoperative > 48-hour, preoperative ≤ 48-hour, postoperative), and univariate analysis showed that incidence rate was highest in the postoperative group (11.8%) and lowest in the preoperative ≤ 48 h group (2.0%). On the other hand, multivariate analysis showed that timing did not have a statistically significant effect, though ICU admission and tube feeding were identified as independent risk factors. CONCLUSION: To increase the effectiveness of intervention, optimization of planning, particularly timing, and accumulation of results for identifying high-risk patients are required. This study indicates that, when evaluating the effectiveness of perioperative oral care, it is necessary to consider structural factors related to disease severity rather than focusing solely on the timing of intervention, and that these factors should be incorporated into future considerations of perioperative oral care practice.