Discussion

This multicenter study revealed that antimicrobials were prescribed in ∼17% of all dental care occasions, with APPs and APTs accounting for 80% and 20% of the prescriptions, respectively.

Antimicrobial misuse was identified in both APPs and APTs. Approximately 10% of APPs were for procedures that did not require antimicrobial prophylaxis. Moreover, even when an APP was indicated, most prescriptions were inconsistent in terms of timing and dosage; only 20% of the patients received a preprocedural APP only, and amoxicillin was rarely prescribed at the standard dosage of 2 g. Although there is no consensus on evidence-based APP in dentistry, ^{Reference Thompson, Teoh and Hubbard19} the guidelines of professional societies and a systematic review have suggested that amoxicillin 2–3 g only at the preprocedural timing should be considered the standard regimen ^{Reference Esposito, Grusovin and Worthington10,Reference Lodi, Figini, Sardella, Carrassi, Del Fabbro and Furness11} and have discouraged postprocedural antimicrobial prophylaxis. ^{Reference Allegranzi, Bischoff and de Jonge20} Although the reasons for the inappropriate timing of APPs observed in the present study are unclear, various factors, including the prescribing practices in the workplace, previous education, and inexperience in prophylaxis using higher dosages of amoxicillin at the pre-procedural-only timing, might be associated with inappropriate prescribing behaviors. ^{Reference Kano, Tagashira, Kobayashi and Honda21}

Although APT accounted for a relatively small proportion of the total antimicrobial prescriptions, a considerable number of APT (∼30% of the total APT) were considered unnecessary. As noted in the treatment guidelines, odontogenic infections may not always require antimicrobial treatment, and dentists need to determine the need for antimicrobial prescriptions based on clinical data. ¹⁴ Moreover, broad-spectrum antimicrobials, such as macrolides, quinolones, and clindamycin, were prescribed in ∼20% of APT. The misuse of these and other antimicrobials has a number of deleterious effects, including undermining antimicrobial stewardship efforts, promoting antimicrobial resistance, and developing antimicrobial-related adverse drug events. ^{Reference Gross, Suda and Zhou22}

Another notable finding was evidence of significant variations in APP and APT prescribing practices at the participating institutions. (Supplementary Table 3). The lack of a standardized treatment for odontogenic infections might be contributing to this marked variation in antimicrobial prescribing practices. Moreover, a previous survey suggested that antimicrobial prescribing practices were strongly influenced by the prescribers’ previous education and workplace culture. ^{Reference Kano, Tagashira, Kobayashi and Honda21} Another study also demonstrated that differences in dentists’ level of knowledge about antimicrobial prescribing practices might be contributing to the variations in their prescribing behaviors. ^{Reference Stein, Farmer, Singhal, Marra, Sutherland and Quinonez23} These findings indicate that the current education on antimicrobial stewardship in dentistry is insufficient. ^{Reference Teoh, Thompson and Suda24–Reference Holz, Naavaal, Stilianoudakis, Carrico, Byrne and Myers27} For effective antimicrobial stewardship, establishing guidelines for standard APP and APT practices is urgently needed. Educational opportunities for dentists to learn about the indications for antimicrobial prophylaxis per type of dental procedure and for antimicrobial treatment of odontogenic infections is the first necessary step toward establishing antimicrobial stewardship in dentistry.

This study had several limitations. Because of its cross-sectional study design, the findings of the present study may not represent long-term trends in antimicrobial prescribing practices in Japan. Second, the findings may not be generalizable to other healthcare settings, such as private clinics, which account for 80% of dental treatment facilities in Japan. ²⁸ Moreover, factors associated with the misuse of antimicrobial prescriptions were not identified in this study. Determining the appropriateness of antimicrobial prescriptions through an EMR review might have introduced a bias due to the limited information contained in the records. The duration of antimicrobial prescriptions was not tracked because only the first instance of antimicrobial prescription was considered. The incidence of adverse drug events due to antimicrobial use may also have been underestimated because patients might have visited other institutions after experiencing symptoms.

In conclusion, the 2 categories of antimicrobial prescribing practice in dentistry examined by this study were APP and APT, both of which were common. Although we did not examine the differences between these 2 patterns in detail, the characteristics of misuse occurring in each differ significantly. The main problems in APPs consisted of inappropriate dosing and timing whereas in APTs the main problems were a high proportion of unnecessary antimicrobial prescriptions and the excessive use of broad-spectrum antimicrobials. The lack of well-established evidence in dentistry for prophylactic and therapeutic antimicrobial practices further complicates the situation. Because significant quantities of antimicrobials are prescribed by dentists, optimizing antimicrobial prescribing practices in dentistry is crucial. Our findings suggest that a better approach to using antimicrobials for APPs and APTs is urgently needed.