Abstract Aim: we sought to estimate the association between hypoglycemic medications especially sodium-glucose co-transporter-2 inhibitors (SGLT2i) and osteomyelitis based on the FDA adverse event reporting system (FAERS). Methods: Publicly available FAERS data were analyzed using reporting odd ratio (ROR) method and Bayesian confidence propagation neural network (BCPNN) method. The developing trend of ROR were revealed by series of calculation based on accumulating dataset quarter by quarter. Results: Ketoacidosis, infections, peripheral ischemia, renal impairment, inflammation including osteomyelitis might more likely to occur among SGLT2i users, especially canagliflozin. Osteomyelitis and cellulitis are AEs unique to canagliflozin. Among 2,888 osteomyelitis-related reports referring to glucose lowering medications, 2,333 cases were associated with SGLT2i, mostly with canagliflozin counting 2,283 which generated an ROR value of 360.89 and a lower limit of information component (IC025) of 7.79. No BCPNN-positive signal could be generated for drugs other than insulin, canagliflozin or drug groups excluding canagliflozin. Reports referring to insulin could generate BCPNN-positive signals during the entire timespan from 2004 to 2021, while BCPNN-positive signal emerged since second quarter (Q2) of 2017, four years since the approval of SGLT2i in Q2 of 2013, for canagliflozin and drug groups containing canagliflozin. Conclusion: This data mining revealed that strong association between canagliflozin treatment and developing osteomyelitis which might be a precursor to lower extremity amputation. Further study with updated data is needed to better characterize the risk of osteomyelitis associated with SGLT2i.

Objectives This study aimed to measure the prevalence of potentially inappropriate medications (PIMs) in Chinese elderly population according to the 2019 Beers Criteria and to evaluate the associative factors of PIMs. Methods Prescriptions of patients aged over 65 years who were hospitalized during January 2017 to December 2017 were retrieved from Hospital Prescription Analysis Cooperation Project, covering seventy-six facilities in six cities in China. PIMs were identified by 2019 Beers Criteria. Results In total, 40.1% elderly patients were treated with at least one PIM independent of their diagnoses or conditions according to the 2019 Beers Criteria. The most frequently prescribed inappropriate medications independent of diagnoses or conditions were proton-pump inhibitors (12.9%). Diuretics (8.2%) and central nervous system medications (6.9%) also showed a relatively high prevalence among the investigated patients. Based on the Beers Criteria’s lists of drug-disease interactions, elderly patients with central nervous system conditions (including delirium, dementia or cognitive impairment) or with a history of falls or fractures were most commonly exposed to PIMs. Logistic regression revealed that inappropriate medication use was associated with age (≥80 years old), number of conditions (≥5), number of medications prescribed (≥10) and class of healthcare facilities. Conclusion This study was able to confirm that the phenomenon of prescribing potentially inappropriate medications was common in China. Our findings support the importance of developing explicit criteria to detect PIMs in China and implementing effective interventions to promote better quality of prescribing for elderly patients.