Harm Reduct J. 2026 Jun 20. doi: 10.1186/s12954-026-01479-x. Online ahead of print.
ABSTRACT
BACKGROUND: People who inject drugs (PWID) are at increased risk of infective endocarditis (IE). While opioid injection is a major risk factor for IE, the risk associated with specific prescription opioids, such as hydromorphone, remains unclear and is inconsistently reported, despite a hypothesized pathway whereby the insolubility of controlled-releaseformulations may promote injection equipment reuse and bacterial contamination. Therefore, we estimated and compared the 1-year incidence rates of IE among PWID reporting injection of hydromorphone versus non-hydromorphone opioid medications and versus other opioids/drugs.
METHODS: Baseline questionnaire data on PWID in the Ontario Integrated Supervised Injection Services cohort (November 2018-March 2020) were linked to health administrative databases. IE incidence was calculated over 1-year post-baseline. Participants were followed until first of IE, death, or 1-year censoring. Exposure was self-reported recent (past 6 months) injection of: hydromorphone, non-hydromorphone opioid medications of interest, or other opioids/drugs. Cox regression estimated associations, adjusting for age, gender, ethnicity, injection drug use, and prior injection-related infections.
RESULTS: Among the 514 participants included, 191 reported recent injection of hydromorphone, 32 non-hydromorphone opioid medications of interest, and 291 other opioids/drugs. IE incidence rates were low across groups: 1.6, 3.3, and 1.8 per 100 person-years. Adjusted hazard ratio comparing recent hydromorphone injection versus non-hydromorphone opioid medications of interest was 0.64 (95% confidence interval (CI) 0.06-6.53); versus other opioids/drugs was 0.82 (95% CI 0.19-3.51).
CONCLUSIONS: In this cohort of PWID, no clear association was found between hydromorphone injection and IE when compared to non-hydromorphone opioid medications or to other opioids/drugs. Due to low event rates and the resulting wide CIs, the findings are limited by low statistical precision. Further research with larger cohorts is warranted to quantify IE risk of specific injected drugs among PWID.
PMID:42321840 | DOI:10.1186/s12954-026-01479-x
