Articles You Might Have Missed

Article Title #1: Takemoto E, Bolton A, Goetz CT. Inequities in naloxone administration among fatal overdose decedents by race and ethnicity in Pennsylvania, 2019–21. Addiction 2024;119:8:1400–1409.

Background: Pennsylvania has been disproportionately affected by the opioid epidemic with the fourth highest number of overdose deaths in a US state in 2020. Naloxone reduces mortality in opioid overdose and until recently was only available via prescription. Prior studies in other states showed differences in provision of substance use resources and access to naloxone between various socio-economic groups, suggesting that members of minority groups may have less access to this life-saving medication.

Research Question: Does naloxone administration differ by overdose decedent race and ethnicity in Pennsylvania?

Methods: This was a single state, retrospective, descriptive review based on information collected from the State Unintentional Drug Overdose Reporting System (SUDORS), Pennsylvania’s coroners and medical examiners (CME) offices, and the Pennsylvania Department of Health (DOH) Bureau of Health Statistics and Registries (BHSR). Cases were identified using ICD-10 cause of death codes X40–X44 and Y10–Y14 for three consecutive years (2019–2021), and a manual review of files was performed. Drug overdose deaths were defined as overdoses from illicit, prescription, or over-the-counter drugs. Racial and ethnic groups were determined from death certificates as: Non-Hispanic Black, Non-Hispanic White, Hispanic, or Other (American Indian/Alaska Natives, Asian/Pacific Islander and Other). Cases with race/ethnicity recorded as “Other” were excluded from analysis. Twenty-two of Pennsylvania’s 67 counties do not participate in SUDORS and were excluded from the study. Additional exclusion criteria included individuals without coroner/medical examiner reports (8.5% of identified overdose deaths), those without recorded race/ethnicity, and cases with toxicology testing positive for both buprenorphine and naloxone without indication of naloxone administration. The primary outcome was evidence of naloxone administration. Multivariate regression models were employed to estimate the odds of naloxone administration using SAS. Analysis was adjusted for age, sex, education level, marital status, population density in county of death, opioid overdose as cause of death, overdose at home, bystander presence, prior overdose, mental health diagnosis, current treatment for opioid use disorder (OUD), recent institutional release, and season of death (due to seasonal variations in death rates). A total of 11,066 overdose deaths were included in analysis (74% of all overdose deaths during the study period).

Results: Throughout all three years, Black decedents were the least likely to have evidence of naloxone administration at the time of fatal overdose of any type, while White decedents were the most likely. Overdose death rates (per 100,000 population) were highest among Black decedents and increased over the study period (4.3 in 2019, 6.1 in 2020, and 6.5 in 2021). Throughout all study years, Black decedents were less likely to have received naloxone compared to White decedents when adjusted for overdose type and demographic characteristics [2019: odds ratio (OR) = 0.7, 95% confidence interval (CI) = 0.5–0.9; 2020: OR = 0.5 (0.4–0.7); 2021: OR = 0.6, (0.5–0.8)]. Hispanic decedents had similar odds of naloxone administration compared to White decedents. Black decedents appeared to have had lower rates of receiving treatment for OUD, history of a prior overdose, or having a mental health diagnosis. Black decedents also had higher percentages of death from a stimulant without opioid involvement (16.4–23.8%) compared to Hispanic (5.5–6.9%) and White decedents (7.1–8.1%).

Conclusion: Black people had higher rates of death from overdose of any type and lower odds of naloxone administration compared to White and Hispanic people in Pennsylvania during the study period.

Critique: Overall, many deaths may have been missed, or excluded, and demographic data could not be confirmed. The data was limited to the SUDORS database and excluded 18% of overdose deaths during the study period. Race/ethnicity was subject to misclassification and could not be confirmed, and cases identified as “Other” were excluded, introducing potential bias. The use of naloxone among decedents was not confirmed; likewise, local access to naloxone at the location of death was unknown. Additionally, the use of naloxone among decedents maintained on buprenorphine-naloxone may have been missed by this analysis.

Implication for Toxicologists: Toxicologists should advocate for improved equity in access to substance use treatment and ensure that minority groups are not further stigmatized during the treatment and prevention of substance use disorders.

Article Title #2: Kimmel SD, Walley AY, White LF, et al. Medication for opioid use disorder after serious injection-related infections in Massachusetts. JAMA Network Open. 2024;7(7);e2421740.

Background: Opioid use via injection is associated with significant morbidity and mortality including serious injection-related infections (SIRIs). Hospitalization for SIRIs presents an opportunity to initiate medication for opioid use disorder (MOUD, i.e., buprenorphine, methadone, or naltrexone), which has been associated with decreased rates of opioid overdose and all-cause mortality.

Research Questions: How many people with OUD received any type of MOUD within 12 months of hospitalization for SIRI? What were the variables associated with and rates of ongoing MOUD treatment at various time points in the 12-month period following hospitalization for SIRI? What were the re-hospitalization and death rates following SIRI?

Methods: This was a retrospective cohort study of the Massachusetts Public Health Data Wearhouse of individuals aged 18 to 64 years of age with OUD and hospitalization for a SIRI between July 1, 2014 and December 31, 2019. The index hospitalization was the first admission for a SIRI, which included the diagnoses of endocarditis, osteomyelitis, epidural abscess, septic arthritis, and bacteremia. Evidence of OUD and MOUD prescriptions for six months prior to the index hospitalization were included in analysis. Data concerning insurance status, alcohol and drug use, and behavioral health diagnoses (i.e., depression or anxiety) were also recorded. Modified Elixhauser Comorbidity index scores were calculated.

Results: A total of 17,212 cases of SIRI and OUD were identified. Of these, 8,769 cases were analyzed, with mean age of 43.2 years (standard deviation 12), and 57.8% were men. In the 12 months after index hospitalization for SIRI, 4,305 (49.1%) received MOUD during at least one week and the mean number of discrete MOUD initiations was 3.0 (SD 1.7). Of those receiving MOUD, 22.2% received treatment for at least 80% of the follow-up time. There were 5,919 (67.5%) rehospitalizations, 395 (4.5%) incarcerations, and 973 (11.1%) deaths (193, 2.2% from opioid overdose) within a year. In the 3 months after index hospitalization, 48.2% of individuals were re-hospitalized and 3.9% died of any cause. At 12 months after discharge, 1,876 (out of 7,421 cases; 25.3%) individuals received MOUD treatment: 9.3% buprenorphine, 15.6% methadone, and 0.4% extended-release naltrexone. In adjusted models, variables positively associated with any MOUD during the study period included previous prescription(s) for buprenorphine (adjusted odds ratio [aOR] 16.51), methadone (aOR 28.46), or extended-release naltrexone (aOR 2.05), diagnosis of endocarditis (aOR 1.88) or septic arthritis (aOR 1.56) compared to bacteremia, having Medicaid (aOR 1.66), previous naloxone prescription (aOR 1.3), and experiencing homelessness (aOR 1.60). Variables negatively associated with MOUD during the 12 months following SIRI hospitalization included age 50–64 years (aOR 0.55), a filled opioid prescription within six months (aOR 0.54), non-Hispanic Black race (aOR 0.66), comorbidity index score 3 or higher (aOR 0.70), discharge home with services (aOR 0.67), discharge to skilled nursing or rehab facility (aOR 0.78), and patient-directed discharge (aOR 0.80) compared to routine discharge.

Conclusion: Among this cohort, only half (49.1%) of patients hospitalized with a SIRI received MOUD during the following year.

Critique: Limitations included the reliance on ICD and billing codes to attribute SIRIs to OUD, exclusion of MOUD received during hospitalization or incarceration, and omission of data regarding patients who were offered but refused MOUD during their index SIRI hospitalization.

Implication for Toxicologists: Toxicologists (and all providers) must use every encounter with OUD patients as an opportunity to initiate MOUD. There is need for increased advocacy for improved MOUD use and equity prior to hospital discharge and in other (e.g., skilled nursing and rehab) settings.

Article Title #3: Dillon DG, Montoy JCC, Nishijima DK et al. Naloxone and patient outcomes in out-of-hospital cardiac arrests in California. JAMA Network Open. 2024;7(8):e2429154.

Background: Out of hospital cardiac arrests (OHCA) are occurring more frequently and continue to have a poor prognosis, with almost 90% being fatal. An increased proportion of OHCA are suspected to be due to drug overdoses, including opioids. Opioid associated OHCA (OA-OHCA) is theorized to have a unique pathophysiology, where hypopnea leads to hypoxia-driven arrest over a longer period compared to arrhythmia or ischemic events. Naloxone use during OA-OHCA affects respiratory drive and may reverse opioid-related myocardial depression.

Research Question: What is the association between pre-hospital naloxone, before vascular access, during OHCA with non-shockable rhythms and patient outcomes?

Methods: This was a retrospective cohort study of nontraumatic OHCA cases from three Northern California counties between 2015 and 2023. Cases involving patients < 18 years of age or missing medication-related data were excluded. The use of pre-hospital medications (including naloxone), resuscitation efforts, and patient demographics were reviewed. The primary outcome was survival to hospital discharge and the secondary outcome was sustained (> 20 min) of return of spontaneous circulation (ROSC). Determination of drug related OHCA was made by the treating EMS providers. Data were analyzed using logistic regression models and underwent two propensity score models to reduce the risk of selection bias. Additional analyses included average treatment (i.e., naloxone) effect on the treated. Odd ratios (ORs) and 95% confidence intervals (CI) were determined.

Results: A total of 8,339 cases were identified and 8,195 were included for analyses. Of these, 715 (8.7%) were identified as presumed drug-related OHCA and 6,707 (81.8%) to have had non shockable rhythms. Naloxone was administered in 1,165 (14.2%) of OHCA cases, which were more likely to involve younger, male patients with fewer comorbidities. Naloxone administration was associated with presumed drug-related OHCA (OR 8.40; 95% CI, 6.93–10.10) and non-shockable rhythms (OR 1.59; 95% CI, 1.30–1.96). The naloxone group was more likely to achieve ROSC and survive to hospital discharge (34.5% vs 22.9% and 15.9% vs 9.7% respectively). This relationship was maintained when employing nearest neighbor and inverse probability-weighted regression analysis. Naloxone was associated with increased ROSC with absolute risk difference (ARD) using nearest neighbor propensity matching (ARD, 15.2%; 95% CI, 9.9%−20.6%) and inverse propensity–weighted regression adjustment (ARD, 11.8%; 95% CI, 7.3%−16.4%). Naloxone was also associated with increased survival to hospital discharge using both adjustments: ARD 6.2% (CI, 2.3%−10.0%) and ARD 3.9% (CI, 1.1%−6.7%). These data yielded numbers needed to treat of 9 for ROSC and 26 for survival to hospital discharge.

Conclusions: Prehospital naloxone administration during OHCA was associated with increased rates of ROSC and survival to hospital discharge.

Critique: This was a well-designed study, but limited by bias in the definition of drug-associated OHCA and lacked data concerning the involved drug(s), with the naloxone group having younger, healthier patients. Furthermore, naloxone administration was not characterized in terms of dosing, route of administration or time since OHCA.

Implication for Toxicologists: Naloxone may improve survivability in prehospital cardiac arrest care. These findings should be considered when creating EMS protocols, data reporting strategies, and continuation of resuscitation efforts in the Emergency Department.

Article Title #4: Degenhardt L, Clark B, Macpherson G, et al. Buprenorphine versus methadone for the treatment of opioid dependence: a systematic review and meta-analysis of randomized and observational studies. Lancet Psychiatry. 2023;10:386–402.

Background: Opioid dependence is associated with substantial health and social burdens. Medications for opioid use disorder (MOUD) are effective in improving multiple outcomes for patients; methadone and buprenorphine are the most common MOUD and are both recommended as first-line agents for opioid dependence. Adherence and retention have previously been identified as important outcomes. This study aims to report the availability and data on a comprehensive range of outcomes in opioid dependence and is the first to do so for both RCT and non-RCT studies.

Research Question: How does buprenorphine compare with methadone for opioid dependence on variable outcomes and adverse effects?

Methods: This was a systematic review and meta-analysis, using established guidelines of several databases (MEDLINE, Embase, CENTRAL, and PsycINFO) from inception to August 1, 2022. Unpublished and ongoing studies were also searched for in other databases (e.g., ClinicalTrials.gov). All studies comparing buprenorphine and methadone, and research on retention of buprenorphine therapy, were identified. Included studies were clinical and randomized controlled trials (RCT), cohort, and case–control studies. Studies were excluded if they involved case reports, case series, cohorts not receiving buprenorphine, studies without relevant outcomes, non-original data, studies involving fewer than 50 subjects, and conference abstracts reported elsewhere. Primary outcomes included retention in opioid agonist treatment at 1, 3, 6, and 12 months, adherence to opioid agonist treatment, and extra-medical opioid use. Secondary outcomes involved use of other drugs, opioid craving, precipitated withdrawal, criminal activity/engagement with the criminal justice system, mental health, non-fatal overdose, physical health, sleep quality, pain, global functioning/treatment satisfaction, adverse events.

Results: The combined search identified 101 studies comparing methadone and buprenorphine (32 RCTs and 69 observational) and 175 studies reporting on treatment retention with buprenorphine (51 RCTs and 124 observational) including a total of 1,030,827 participants from around the world. For RCTs and non-RCT studies, retention in treatment did not differ significantly between methadone and buprenorphine at one month. Beyond the first month of treatment, retention was consistently better for methadone. At six months, RCTs (pooled risk ratio [prr] 0.76 [95% CI 0.67–0.85]), and non-RCT studies (prr 0.77 [0.68–0.86]) showed similar retention rates with methadone (56%) compared to buprenorphine (52%). Retention was generally better in RCTs compared to observational studies. There was evidence of reduced extra-medical opioid use in those receiving buprenorphine in RCTs when measured via urinalysis (standardized mean difference −0.2 [−0.29 to −0.11]). There was no significant difference in this outcome when using other measures. There was limited evidence of increased treatment satisfaction and reduced cocaine use, cravings, anxiety, and cardiac dysfunction with buprenorphine, and reduced alcohol use and hospitalization with methadone. There were no significant differences in adverse effects from either treatment.

Conclusion: Evidence from clinical trials and observational studies suggest that treatment retention is marginally better with methadone compared to buprenorphine after one month. Overall retention at and beyond 6 months was relatively poor for both buprenorphine (52%) and methadone (56%). Secondary outcomes suggested statistically better outcomes for buprenorphine, including satisfaction, cravings, anxiety, and cardiac dysfunction.

Critique: Although this paper utilizes a large dataset encompassing studies across the globe, there were significant limitations and biases. Inconsistent methods and measures limited the ability to pool results. Primary outcomes were subject to bias from randomization and confounding factors. Secondary outcomes were analyzed using a small number of inconsistent studies.

Implication for Toxicologists: These data show the importance of identifying and engaging variables to increase retention rates and duration for treatment with MOUD. Low retention rates should be recognized to improve expectations and encourage future research.

Declarations

Conflicts of Interest

None.

Sources of Funding

None.

Competing Interests

The authors have no competing interests.